A Review of Nonaqueous Electrolytes, Binders, and Separators for Lithium-Ion Batteries

Abstract

Lithium-ion batteries (LIBs) are the most important electrochemical energy storage devices due to their high energy density, long cycle life, and low cost. During the past decades, many review papers outlining the advantages of state-of-the-art LIBs have been published, and extensive efforts have been devoted to improving their specific energy density and cycle life performance. These papers are primarily focused on the design and development of various advanced cathode and anode electrode materials, with less attention given to the other important components of the battery. The “nonelectroconductive” components are of equal importance to electrode active materials and can significantly affect the performance of LIBs. They could directly impact the capacity, safety, charging time, and cycle life of batteries and thus affect their commercial application. This review summarizes the recent progress in the development of nonaqueous electrolytes, binders, and separators for LIBs and discusses their impact on the battery performance. In addition, the challenges and perspectives for future development of LIBs are discussed, and new avenues for state-of-the-art LIBs to reach their full potential for a wide range of practical applications are outlined.

Graphic Abstract

1 Introduction

There is a growing demand for low-cost, large-scale energy storage of electricity generated from intermittent renewable energy sources, such as solar panels and wind turbines, to support various stationary and transportation applications. Lithium-ion batteries (LIBs) are recognized as the most advanced energy storage devices for these applications because of their high energy density, high power density, longer cycle life, and higher cell voltage in comparison with other secondary batteries [1,2,3]. Despite these advantages and the commercial viability of rechargeable LIBs, some challenges remain, such as poor rate capability, large volume change during charging and discharging, insufficient energy density, and safety issues that must be overcome to enable a broader range of applications [4]. In fact, reduced cost, improved safety, and enhanced energy density are mandatory requirements for full commercialization of LIBs [5, 6]. Some of these issues have been addressed, and Tesla’s electric vehicles (EVs) have had significant success using LIBs for storing electricity for long-range (300–400 miles per charge, 1 mile = 1 609.344 m) travel. For vehicle applications, power is most important, and the materials must be able to charge sufficiently fast to take advantage of regenerative braking. There is still significant room for improvement in battery range, charging rate, lifetime, and safety.

Currently, LIBs rely on graphite anodes, lithium metal oxide cathodes, and liquid organic carbonate electrolytes [7]. A large portion of the research on LIBs has focused on the modification of the morphology of electrode materials, such as surface coating, dimension reduction, porosity engineering, doping, and functionalization [8,9,10]. These explorations are very important for future LIB development, but currently, they are still in the research and development stage and are not ready for commercialization [11]. The well-accepted cathode and anode materials in the industry are still lithium metal oxide and graphite [12]. The research interest in developing high-capacity electrode materials remains high, and new materials that will increase the specific energy density of LIBs are under development.

In recent years, it has become more evident that nonelectroconductive components of batteries, such as electrolytes, binders, and separators, have a significant impact on the electrochemical performance as well as safety issues that LIBs face [3]. In addition, the development of the nonelectroconductive components of the battery is recognized to be more cost-effective and economically preferable. Hence, developing and understanding the nonelectroconductive factors in LIBs, such as electrolytes, binders, and separators, can guide the design of LIBs and improve their performance and reliability [13]. The importance of the nonelectrode components for the development of next-generation LIBs is acknowledged in the literature, and some review papers have focused primarily on the advancements in the development of individual nonelectrode components of LIBs [14,15,16,17,18,19]. This review paper is specifically focused on the recent developments in all nonelectroconductive components of LIBs.

The progress in the development of electrolytes for LIBs is discussed in the second section of this review. The electrolyte is a critical and essential element of the nonelectroconductive components. It can impact the battery energy density, rate capability, interfacial chemistry, and other key battery performance factors [20]. The development of novel functional electrolytes based on solvent mixtures, salts, and functional additives has the primary objectives of enhancing electrode passivation, removing harmful contaminants, reducing flammability, and achieving both overvoltage protection and fast charging [21]. Figure 1a shows the historical development of the electrolytes for LIBs. Liquid electrolytes continue to be the most common electrolyte used in batteries today. The current state-of-the-art liquid electrolyte is composed of a lithium salt in a mixture of ethylene carbonate (EC) and a linear carbonate such as dimethyl carbonate (DMC). This mixture provides high conductivity and favorable electrode passivation against parasitic side reactions [22]. However, safety issues associated with these liquid electrolytes are still unresolved, and further development is required to overcome these concerns [23, 24]. Recently, considerable efforts have been spent on developing gel and solid electrolytes (Fig. 1a) based on polymers [20] and ceramics to meet the safety requirements [25,26,27]. Furthermore, ionic liquids (ILs) are attracting increasing interest as nonflammable and high-voltage electrolytes [28, 29].

Fig. 1

a Development of electrolytes for LIBs. b Typical examples of polymer binders. c Classification of separators for LIBs

The polymer binder and separator are indispensable parts of the battery design. Figure 1b and Fig. 1c list typical examples of polymer binders and separators that are discussed in detail in the third and fourth sections of this review paper, respectively. Polymer binders bond the active material and conductive additives to maintain the integrity of the electrode and ensure the stability of the structure (Fig. 1b) [30]. They also play an important role in preserving the mechanical stability, flexibility, and electroconductive pathways of the electrode. The separator membrane provides an electrical barrier between the two electrodes and serves as the medium for lithium-ion transport (Fig. 1c) [31]. In addition to providing a physical barrier for the electrodes, the separator acts as an electrolyte reservoir that ensures a sufficient number of lithium ions and their mobility during battery operation. Generally, binders and separators are not active components during charging and discharging. However, the properties of the binders and separators directly influence the performance of the batteries.

In the final section of this review paper, the challenges and perspectives for future development of state-of-the-art LIBs are discussed, and potential research directions for future advancement in this field are outlined.

2 Nonaqueous Electrolytes

The electrolyte is an ion-conducting medium that allows ions to move between the cathode and anode. Great progress has been made in the development of electrolytes since the invention of LIBs. The general trend in electrolyte development followed the path illustrated in Fig. 1a. Traditional liquid electrolytes were used predominantly in the first few decades before gel, and polymer electrolytes were developed in the 1990s. The next advancements in electrolyte development were associated with IL electrolytes and, most recently, solid electrolytes.

2.1 Liquid Electrolyte Solutions

Basically, the liquid electrolytes for LIBs are composed of lithium salts, organic solvents, and functional additives. The most important requirements that a good electrolyte must satisfy are as follows: (1) facilitating the formation of a stable passivating layer; (2) possessing limited volume expansion; (3) having high ionic conductivity; and (4) featuring low flammability. Notably, no current electrolyte can satisfy all these requirements.

During the past few years, numerous strategies have been explored to overcome the challenges and facilitate the formation of a stable passivating layer, including (1) tailoring the electrolyte components using overcharge protection additives, materials with positive temperature coefficients, and redox-shuttle additives [32] and (2) using different ratios of solvent mixtures such as LiPF₆⁺ [EC: linear alkyl carbonate (ethyl methyl carbonate (EMC), DMC, diethyl carbonate (DEC), etc.)] rather than a single solvent. Herein, we summarize the recent progress in the development of additives, salts, and solvents. These three components give us a view of the electrolyte design strategies explored in recent years.

2.1.1 Additives

The development of appropriate electrolyte additives is considered the most feasible, economical, and effective strategy to overcome the major problems hindering the advancement of LIBs as an innovative energy storage technology [33]. An electrolyte additive (usually less than 10% by weight or volume percentage) serves as a sacrificial component used as a foreign molecule to adjust the targeted properties of the interphase so that it will reduce or oxidize before the parent electrolyte without changing the composition framework. In the past few decades, research on electrolyte additives has far surpassed the development of new electrolytes.

The highest occupied molecular orbital (HOMO) energy has proven to correlate to the potential of the oxidative decomposition reaction on the electrodes, and the lowest unoccupied molecular orbital (LUMO) energy has proven to correspond to the potential of the reductive decomposition reaction [34]. Therefore, one important requirement for additives is the difference between the LUMO and HOMO, which is the value of the electrochemical stability window (E_g) of the electrolyte [35]. Ideally, the E_g of the electrolytes needs to be wide enough to cover the redox potential of the anode and cathode to avoid decomposition reactions on the electrodes. In this case, a higher LUMO or a lower HOMO will be preferred for a sufficiently wide E_g. The value of E_g for each key component in the battery has an intrinsic effect on the battery chemistry. To obtain a high-quality interface for the anode and cathode, the additives must be reduced prior to the electrolyte components. In this way, the additives can passivate the negative electrode prior to any reductive decomposition of the electrolyte. Electrolyte additives are pivotal for stabilization of LIBs, which they achieve by suppressing capacity loss through the creation of an engineered solid–electrolyte interphase layer (SEI layer) at the negative electrode [36].

Electrolyte additives with various functional groups are widely used in LIBs [37]. The functional groups include inorganic and organic compounds containing unsaturated carbon bonds, sulfur-containing components, halogen-containing components, and other components [38]. For better battery performance, additives can (1) facilitate the formation of an SEI on the surface of the anode and cathode, preventing electrolyte decomposition, (2) reduce the irreversible capacity and gas generation for SEI formation and long-term cycling, (3) enhance the thermal stability of LiPF₆ against the organic electrolyte solvents, (4) protect the cathode material from dissolution and overcharge, and (5) improve the physical properties of the electrolyte such as the ionic conductivity, viscosity, and wettability to the polyolefin separator [5].

For better battery safety, additives can (1) lower the flammability of organic electrolytes, (2) provide overcharge protection or increase the overcharge tolerance, and (3) terminate battery operation during abnormal operating conditions [33]. Herein, we review two types of additives reported in recent publications and discuss their principal functions.

2.1.1.1 Additives Impacting the SEI Formation in the Anodes

Additives Designed to Stabilize, Modify, and Enhance the SEI Formation in the Anodes An effective electrolyte additive for the anode needs to be reduced before the electrolyte is reduced, and it needs to form a passivating film. Moreover, the additive should prevent the continuous growth of the SEI. In particular, an SEI-free solid electrolyte is crucial for a high initial Coulombic efficiency and fully reversible conversion [39]. The most commonly used electrolyte additive is vinylene carbonate (VC), which undergoes polymerization at the negative electrode surface, creating a protective SEI layer. However, VC has a negative impact on performance due to oxidation at the positive electrode, especially at elevated temperatures [40].

Sulfone compounds have also been studied as effective SEI-forming additives. For example, propane-1,3-sulfone (PS) [41] and prop-1-ene-1,3-sultone (PES) [42] were both proven to stabilize the SEI layer. Some of these sulfur-containing SEI-forming additives have been studied by density functional theory (DFT) calculations to elucidate the reduction mechanism [5, 43]. Research has shown that Li⁺ ions play a pivotal role in the reduction decomposition of PES [44]. The decomposition processes of O–C and S–C bond breaking lead to stable decomposition products, while the kinetically favored S–O bond breaking process is blocked by the formation of a Li⁺-participated seven-membered ring, which were identified as RSO₃Li and ROSO₂Li and were observed in the SEI layer [45, 46].

Nitrogen-containing additives such as nitrile [47] and nitrogen-containing compounds [48] may be more likely to attract an electron and be reduced on the anode compared with carbonate-based electrolytes. Ji et al. conducted X-ray absorption near-edge spectroscopy (XANES) studies and found that charge transfer from suberonitrile (SUN) to LiCoO₂ results in the formation of a lower valence of Co⁴⁺, which stabilizes the electrode/electrolyte interface [49].

In addition, several other methods have been used to improve the quality of the SEI layer. The surface crystallographic planes of the active material play an important role in the composition and structure of the formed SEI layer. In Lyn’s work [50], β-Sn nanorods (NRs) with (200) facets and different aspect ratios were used as anode materials. As-prepared anodes showed good cycling stability at a 0.2 C charge–discharge rate, and high discharge capacities of ∼ 600 and 550 mAh g⁻¹ after 100 cycles were reported for anodes prepared from NRs with high and low aspect ratios, respectively (Fig. 2a, c). According to the authors, the improvement is due to the combination of the NR morphology that buffers large volumetric changes and the LiF-rich F-containing SEI layer that allows for a stable SEI layer and good ionic and electrical conductivity [50]. In another study, Fulu et al. proposed a series of metal–organic framework (MOF) particles as new potential solid additives to inhibit Li dendrite formation at the anode–electrolyte interface (Fig. 2b, d) [51].

Fig. 2

Copyright © 2019, American Chemical Society. b X-ray diffraction (XRD) patterns and scanning electron microscopy (SEM) images of UiO-66. Reproduced with permission [51]. Copyright © 2019, American Chemical Society. c Cyclic test of high and low aspect ratio β-Sn NRs and spherical β-Sn electrodes. Reproduced with permission [50]. Copyright © 2019, American Chemical Society. d Li plating/stripping performance comparison for the cases free of additives and containing 1.0 wt% UiO-66 grains at 1 mA cm^–2 with a capacity of 3 mAh cm^–2. Reproduced with permission [51]. Copyright © 2019, American Chemical Society

a Selected area electron diffraction (SAED) analysis of β-Sn NRs indicating the growth direction and surface planes, and type of SEI layer compound formed on the surface of β-Sn NRs with respect to the surface crystallographic plane. Reproduced with permission [50].

Additives for High-Capacity Anodes Such as Silicon and Li Metal The high-energy–density application of LIBs is limited by the capacity of the graphite anode [52]. Silicon is a promising anode material that may be able to replace graphite as an active electrode material. Si-based electrodes have high theoretical capacities (nearly 4 200 mAh g⁻¹) compared to carbon-based electrodes (372 mAh g⁻¹) [53, 54]. However, Si anodes have a large volume change (over 300%) during charging and discharging, which results in a decrease in both the cycle life performance and the stability of the battery [55]. One solution for this problem is the use of electrolyte additives.

Fluoroethylene carbonate (FEC) [56] and VC [56] are the most commonly used electrolyte additives for both Si- and Sn-based electrodes. With Si electrodes, FEC is used as a cosolvent, which will be covered later. Hy et al.’s work showed that through a weblike network formed by combining the VC electrolyte additive with an Al₂O₃ atomic layer deposition (ALD) coating, the cycling stability of the Si/reduced graphene oxide (rGO) product/carbon nanotube (CNT) composite anode was dramatically improved (73% to 11% loss after 50 cycles) [55] (Fig. 3a, c). Another additive reported to improve the performance of Si anodes is succinic anhydride (SA) [57]. Research shows that the presence of low concentrations of SA helps enhance the formation of Li₂CO₃ and hydrocarbons, which prevent the decomposition of LiPF₆. It also increases the Coulombic efficiency, cycling performance, and capacity retention of the battery.

Fig. 3

Copyright © 2015, American Chemical Society. b Schematic of selective ALD LiF deposition on h-BN. Reproduced with permission from AAAS [55]. c Charging capacity versus cycle number and charging capacity versus current density on a log scale showing the rate capability of the bare and 20 ALD cycle electrodes with and without the VC additive. Reproduced with permission [55]. Copyright © 2015, American Chemical Society. d Inferred electrodeposition mechanism: lithium diffusion near the surface of electrodes represented by downward arrows. The LiF host is shown in green. Due to the lower diffusion barrier on LiF, wafer-like Li deposition is expected, while in the usual SEI, needle-like Li plating is formed. Reproduced with permission [64]. Copyright © 2018, Elsevier B.V

a Diagram of the interaction of ALD and VC during cycling. Reproduced with permission [55].

To improve the electrochemical performance of silicon-based anode materials, lithium fluoride (LiF) was introduced. Yang et al. used pitch carbon and LiF to co-modify a silicon/graphite (S/G) composite. The modified silicon/graphite sample showed a capacity of over 500 mAh g⁻¹, whereas unmodified S/G delivered a capacity of less than 50 mAh g⁻¹ after 100 cycles at 100 mA g⁻¹. When operated at 4 A g⁻¹, the reversible capacity of the modified SG was 346 mAh g⁻¹, much higher than that of S/G (only 37 mAh g⁻¹). The enhanced cycling and rate properties of the modified silicon/graphite were attributed to the synergetic contributions of the pitch carbon and LiF [58]. In addition, Roy et al. chemically exfoliated a few layers of black phosphorus as a simple physical additive, providing a Si nanoparticle-based anode with very high capacity, power density, and stability [59]. Park et al. synthesized a silicon diphosphide (SiP₂) compound having a 3D crystalline framework with sixfold coordination of Si, in which the Li-ion storage mechanism involves a three-step reaction between SiP₂ and Li. Trimethyl phosphate (TMP) [60] was found to improve the cycling stability and rate capability of Li_1.2Mn_0.54Ni_0.13Co_0.13O₂ up to 4.8 V (vs. Li/Li⁺), which was attributed to the SEI formed by TMP [61].

In another work, Feng et al. designed a LiF-enriched lithophilic 3D matrix on Cu foam with a high specific surface area and a stable protective layer. The LiF@Li matrix exhibited excellent stability for 350 h at a high areal capacity of 5 mAh cm⁻² with a current density of 5 mA cm⁻². The stability is understood to be due to the 3D current collectors, which have more space for Li plating [62].

The lithium (Li) metal anode is among the most promising anodes due to its high energy density; however, the long-standing issue of Li dendrite growth hinders the practical application of lithium metal batteries (LMBs). Choudhury and Archer [55] found that electrolytes containing 0.5 wt% (wt% means the weight ratio percentage) LiF in a 1 M (1 M = 1 mol L^–1) electrolyte solution offer remarkable dendrite suppression abilities and enhanced electrolyte stability during battery cycling. The batteries with the LiF additive showed low interfacial resistance and over 90% Coulombic efficiency due to better protection of lithium metal compared to conventional electrolytes. The authors also reported that the LiF additive enhanced the battery lifetime to hundreds of hours by suppressing dendrite growth. The addition of LiF salt to the electrolyte improves the diffusion of ions on the electrode surface, which leads to smooth electrodeposition. This LiF additive also forms a protective layer on the electrode that prevents the degradation of the electrolyte by side reactions. It can be effective in improving the capacity and lifetime of a commercial LMB [63]. Yuan synthesized a LiF host with good Li⁺ diffusivity and chemical stability to restrain dendrite growth and improve the cycle life performance of LMBs. Not only does the LiF host serve as a chemically stable interfacial layer for Li anodes by inhibiting corrosion reactions, but also the abundance of pores in the LiF host can control the growth of the Li deposits in a flat and scale-like morphology. The LiF-modified LMB displayed stable cycling over 520 h and more than 98.5% Coulombic efficiency at a current density of 1.0 mA cm⁻² [64].

Cui’s group performed selective ALD of LiF on defect sites of hexagonal boron nitride (h-BN) by chemical vapor deposition [57]. The chemically and mechanically stable hybrid LiF/h-BN film successfully suppressed lithium dendrite formation during both the initial electrochemical deposition onto a copper foil and the subsequent cycling (Fig. 3b, d) [64]. The modified lithium electrodes exhibited good cycling behavior with more than 300 cycles at a relatively high Coulombic efficiency (> 95%) in an additive-free carbonate electrolyte [65].

Additives for a 4-V Cathode Such as Spinel LiMn₂O₄ and Olivine LiFePO₄ A key factor that limits the performance of LIBs is the dissolution of cathode materials at either elevated temperature or higher potential, which results in capacity loss and poor cycle life performance. One approach to solve these problems is to use additives that form an effective surface film on the cathode. The selected cathode additives oxidize before their solvents and form a film that will cover the electrode surface and prevent the decomposition of the electrolytes. The formation of a protective surface film by an electrolyte additive does not hinder interfacial charge-transfer reactions. For example, the presence of 2 wt% lithium difluoro(oxalato)borate (LiDFOB) in a LiPF₆/EC:EMC (3:7, v/v, v/v means to be in the volume ratio) solution resulted in 92% retention of the initial capacity after 100 cycles, compared to only 58% without this additive [66]. Similarly, lithium bis(oxalate)borate (LIBOB) and tris(2,2,2-trifluoro ethyl)phosphate also effectively stabilized the capacity and discharge voltage [67]. In another study, Rong et al. [60] showed that 1% tris(trimethylsilyl) phosphate (TMSP) dramatically decreased the impedance of graphite/LiNi_0.4-Co_0.2Mn_0.4O₂ cells, resulting in a significant improvement in the cycling stability (from 57.6% to 90.8% discharge capacity retention) and rate performance. Researchers have also reported discharge capacities of 148.9 and 112.9 mAh g⁻¹ for batteries with TMSP additives at 2 C and 3 C discharge rates, respectively, while the values were only 137.7 and 88.8 mAh g⁻¹ for the cells with blank electrolytes at the same rates [68].

Modified F-LiFePO₄/C cathode materials with LiF additives were synthesized by Gu et al. [69]. Their results showed that the LiF additive lowers the LiFePO₄ formation temperature by approximately 100 °C. Modification with LiF improves the crystallinity of LiFePO₄ and results in a smaller particle size. Charge–discharge curves indicated that the LiF-modified LiFePO₄ cathode has excellent capacity at high discharge rates (130.0 mAh g⁻¹ at 5 °C) [69]. Similar results were obtained by Sun et al. [70]. When LiF is used as an additive to LiFePO₄, the mixing of ultrafine lithium fluoride and metal particles provides a high lithium conversion efficiency. As reported, the LiF/Co nanocomposite exhibits an open-circuit voltage (OCV) of 1.5 V, which is comparable to the OCVs of the existing cathode materials and delivers a high first-cycle “donor” lithium-ion capacity (516 mAh g⁻¹). The LiFePO₄ cathode with the 4.8% LiF/Co additive showed an increased first-cycle charge capacity, ∼ 20% higher than that of its pristine LiFePO₄ counterpart, indicating high prelithiation efficiency. Their work showed the concept of nanoscale mixing of LiF with other metals (transition metals such as Ti, V, Mn, Ni, Cu, and Zn and main group elements such as Al and Sn) that also work as prelithiation additives [70].

Additives for High-Voltage Cathodes High-voltage cathodes, such as lithium-rich cathodes, have high potential (> 4.5 V), high capacity (> 250 mAh g⁻¹), and good thermal stability [71], but they have poor cycle life performance [47, 72]. Phosphorus derivatives have been reported as electrolyte additives for high-voltage cathodes, including TMP [61]. Mai et al. reported a novel electrolyte additive that combines TMP, triphenyl phosphate (TPP), and dimethyl phenylphosphonite (DMPP). The authors reported that this additive significantly improves the cycling stability of the LiNi_0.5Mn_1.5O₄ cathode for high-voltage LIBs at elevated temperatures [73].

Another important type of additive for enhancing the performance of high-voltage cathodes is fluorine-containing compounds. The use of low concentrations of fluorine functional groups is an efficient and cost-effective strategy for improving the electrolyte stability and providing better cycling performance compared to conventional electrolytes. Research shows that di(2,2,2 trifluoroethyl) carbonate (DFDEC) can reduce voltage fade and maintain the structural stability of Li-rich cathode materials [74].

2.1.1.2 Safety Protection Additives

Lower Flammability Additives Phosphorus-containing organic compounds are well known and practical for use as flame-retardant (FR) materials [75] to suppress the flammability of liquid electrolytes in LIBs. The flame retardation mechanism is based on a physical char-forming process and chemical radical scavenging.

As reported by Xu et al., for a conventional LiPF₆-based carbonate electrolyte, a TMSP and 1,3-propanediol cyclic sulfate (PCS) binary functional additive (as shown in Fig. 4) can improve the performance of high-voltage (5-V-class) LiNi_0.5Mn_1.5O₄/MCMB battery systems at both room temperature and 50 °C [34]. The LiNi_0.5Mn_1.5O₄/MCMB battery with the binary functional additive showed preponderant discharge capacity retention of 79.5% after 500 cycles at a 0.5 C rate at room temperature. Additionally, the introduction of (ethoxy)pentafluoro-cyclo-triphosphazene (PFPN) can reduce the flammability of the aforementioned binary functional additive-containing electrolyte [34].

Fig. 4

Copyright © 2017 WILEY–VCH

Chemical structures of the as-prescribed functional additives: a TMSP, PCS, and PFPN. The HOMO and LUMO energy levels of the TMSP, PCS, and PFPN additives and carbonate solvents (EC/DEC/EMC) were calculated. b Discharge capacity of the LiNi_0.5Mn_1.5O₄/MCMB full cells with BE and BE + binary functional additives (1 wt% TMSP + 1 wt% PCS) at various C-rates. c First (top) and second (bottom) charge–discharge voltage curves of the LiNi_0.5Mn_1.5O₄/MCMB full cells with BE and BE + binary functional additives (1 wt% TMSP + 1 wt% PCS) at a 0.2 C rate. Reproduced with permission [34].

Beyond a threshold level, ILs can endow nonflammability to organic solvents. The LIB community clearly cannot ignore this option for the safe operation of batteries [76].

Overcharge Protection Overcharge is one of the abnormal operating conditions for LIBs. Overcharging a battery elevates the battery temperature, which can result in a fire or even an explosion. To prevent the battery from overcharging, the additives should be electrochemically reversible at slightly higher cell potentials than the maximum operating potential of the cathode. Additionally, the oxidation potential of the additives should not exceed the operating electrochemical window of the electrolytes; otherwise, safety concerns may arise.

2,5-Di-tertbutyl-1,4-dimethoxybenzene (DDB) has been reported to be a good additive for LiFePO₄ cathode overcharge protection materials [77]. Because of its robust molecular structure, DDB shows excellent electrochemical stability and reversibility. However, its lower potential (3.9 V vs. Li/Li⁺) limits its application in high-voltage electrode materials. Modifying DDB with more-electron-withdrawing trifluoroethoxy groups [i.e., 1,4-di-t-butyl-2,5-bis(2,2,2-trifluoroethoxybenzene), DBTFB] improves the potential (4.25 V vs. Li/Li⁺) of the additive and its application in cathode materials with potentials of up to 4.2 V such as LiCoO₂ or Li[Ni_1/3Mn_1/3-Co_1/3O₂] [78].

Benzene derivatives (biphenyl and o-terphenyl) and heterocyclic compounds (furan, thiophene, N-methylpyrrole and 3,4-ethylenedioxythiophene) can also be used as additives for overcharge prevention. These additives decompose on the cathode and form very thin films, which are known as overcharge protecting proofs [79].

2.1.1.3 Others

Additives to Develop High Transference Number Electrolytes The transference number for lithium is important for the power and energy density of LIBs. Kyle M. Diederichsen et al. demonstrated that electrolytes with higher lithium transference numbers possess higher power densities and enable faster charging [80]. Conventionally, enhancement of the lithium transference number is realized by using anionic acceptors. However, this approach often suffers from the steric hindrance effect when applied to relatively large supramolecular additives, leading to lower electrolyte conductivity. Recently, some of these limitations were overcome when calix pyrrole (C₆P) [81] was introduced as an anion trapping group. Additionally, anion receptors based on boron compounds [82] were applied to the solution of lithium salts in an inert electrolyte, which dramatically improved the stability of the SEI by dissolving unstable LiF.

Nanocarbon materials, such as CNTs and graphene nanosheets (GNSs), have been extensively studied as conductive additives due to their low cost, reasonably high electrical conductivity, and high durability. Benefiting from their high aspect ratio, which can significantly reduce the volume of conductive additives, the introduction of CNTs/nanofibers significantly contributes to enhancing the transfer number [83,84,85]. Zhang et al. reported the beneficial effects of GNSs with a low percolation threshold on the electrochemical performance of Li₄Ti₅O₁₂ (LTO) anodes [86]. Furthermore, carbon materials can also be used as elastic conducting additives in the form of composites or coatings [87, 88].

2.1.2 Lithium Salts

2.1.2.1 Lithium Fluorophosphate Salts

LiPF₆ is the most prevalent salt used in commercial LIBs based on a graphite anode and a 3-to-4-V cathode. LiPF₆ has a uniquely suitable combination of properties (temperature range, passivation, conductivity, etc.), which can provide the best balance, rendering it the current overall best Li salt for LIBs. However, LiPF₆ is not the ideal Li salt for each important electrolyte property. The chemical and thermal instability of LiPF₆ are the two most common issues. In addition, the use of LiPF₆-containing electrolytes also results in anode performance deterioration [89].

Recent work has shown that LiBF₄ may be a promising alternative salt or co-salt for high-voltage LIBs [90, 91]. LiBF₄ has a similar chemical structure to LiPF₆, and both salts have advantages and disadvantages in lithium-ion cells. LiPF₆ has high ionic conductivity when it is dissolved in carbonate solvents and helps passivate the negative electrode SEI. However, LiPF₆ is known to have poor thermal and hydrolytic stability [92]. There is no electrolyte thermodynamically stable toward each electrode with a voltage of approximately 4 V in LIBs. Interface reactions occur at the interface of electrolytes and electrodes [93]. Gas generation in LIBs is more common than interface reactions between the electrolytes and electrodes. In addition, decomposition of LiPF₆ results in the formation of reactive species, HF and PF₅, which are known to cause a cascade of undesirable side reactions inside the cell [94]. LiBF₄ has often been used as a substitute for LiPF₆ in LIBs because of its improved thermal and hydrolytic stability [95]. Moreover, in some situations, LiBF₄ has proven to be more stable at high voltages. LiBF₄ has demonstrated several other important advantages over LiPF₆, such as improved performance at subzero temperatures, improved passivation of the Al current collector, and improved performance at the positive electrode. Compared to LiPF₆, the disadvantages of LiBF₄ are lower ionic conductivity and poor performance at the negative electrode [96].

Zhang’s work showed that LiBF₄ has improved performance in LIBs at high (50–80 °C) and low (20 °C) temperatures [97]. Additionally, Zhang’s results showed that LIBF₄ provides very good aluminum current collector passivation properties [96].

The accessible capacity, rate capability, thermal reactivity, mechanical stability, and durability of LIBs are largely governed by the porosity and thickness of the SEI layer [89]. The SEI formed during the first charge/discharge cycle due to the instability of the solvent and salt in the electrolyte determines the overall functioning of the battery system. The anion and cation of the electrolyte salt appear to play a key role in determining the overall SEI layer composition, including its depth evolution and thickness. Therefore, adding a binary salt to each solution was investigated to improve the conductivity. Ellis et al.’s work showed that 1.0 M electrolytes consisting of equal parts LiBF₄ and LiPF₆ have better performance in high-voltage (4.35 V) LIBs than electrolytes with LiPF₆ or LiBF₄ alone. They found that an electrolyte with LiPF₆:LiBF₄ (1:1) composition and no other additives performed similarly to an electrolyte with LiPF₆ and state-of-the-art additive blends in these batteries [98].

Many researchers have focused on investigating new lithium salts to replace LiPF₆. Considering performance metrics only, lithium hexafluoroarsenate (LiAsF₆) would certainly be an ideal candidate for an electrolyte Li salt in advanced LIBs. LiAsF₆ provides very good ionic conductivity and excellent cycling performance [99]. Strong As–F bonding results in higher thermal stability [100], increased resistivity to hydrolysis, and enhanced electrochemical stability to oxidation (up to potentials of approximately 4.7 V). However, the possible formation of highly toxic AsF₃ [101] excludes the use of this electrolyte material in commercial LIBs [76].

Another potential alternative to LiPF₆ is lithium tris(pentafluoroethyl) trifluorophosphate (LiFAP), a structurally related derivative of LiPF₆. Compared with LiPF₆, it has improved chemical, electrochemical, and thermal stability as well as an increased resistivity to hydrolysis [102]. Moreover, it shows enhanced performance when employed in high-voltage graphite/LiNi_0.4Mn_1.6O₄ lithium-ion full cells [103].

The search for an alternative salt for the electrolyte in LIBs to attain excellent and competitive performance is ongoing. Jianhui Wang et al. obtained an unusual liquid by mixing the stable lithium salt LiN(SO₂F)₂ with the DMC solvent at extremely high concentrations [1:1.1 bis(fluorosulfonyl) amide (LiFSA)/DMC]. The unusual liquid showed a three-dimensional (3D) network composed of anions and solvent molecules that coordinated strongly with Li⁺ ions [104]. This simple framework of a superconcentrated LiN(SO₂F)₂/DMC electrolyte inhibits the dissolution of both aluminum and transition metals up to 5 V and realizes a high-voltage LiNi_0.5Mn_1.5O₄/graphite battery. The LiNi_0.5Mn_1.5O₄/graphite battery exhibits excellent cycling durability, high-rate capability, and enhanced safety.

Presently, imide-based lithium salts are also among the most promising alternatives to LiPF₆ in commercial devices. Generally, they show excellent chemical, thermal, and electrochemical stability, thereby improving the cycling performance and safety of LIBs. The best-known compound within this family of lithium salts is certainly lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) [105]. The TFSI⁻ anion provides high chemical inertness toward all battery components because it is characterized by a rather low Lewis’s acidity due to the delocalized negative charge [106], which results in superior electrochemical stability to oxidation. In addition, LiTFSI exhibits good ionic conductivity, outstanding solvation ability, very low sensitivity to hydrolysis, and advantageous thermal stability [107]. These characteristics render the LiTFSI salt a highly promising alternative to LiPF₆. If the perfluorinated methyl group for “simple” fluorine atoms is substituted, another very promising alternative lithium salt with the Li[N(SO₂F)₂] (LiFSI) composition can be synthesized [108]. LiFSI does not release significant amounts of highly toxic and corrosive HF under overtemperature conditions and exhibits advantageously high resistivity to hydrolysis.

2.1.2.2 Borate-Based Lithium Salts

The development of a new lithium salt to enhance the interfacial stability between positive/negative electrodes, and electrolytes has been another core topic. The newly developed salts for LIBs include borates and boron-based cluster and phosphate-type salts. There are salts in which anions comprise ligands around a central atom (such as PF₆⁻ and ClO₄⁻) as well as large complex anions such as TSFI [109] and organic ligand-based anions [e.g., bis(oxalate)borate (BOB)] [110]. Lithium orthoborate salts have been intensively studied during the last two decades because of their distinct thermal stability and potential for replacing the LiPF₆ salt in commercialized electrolytes, which has low chemical and thermal stability [106]. One particular member is LiBOB, which shows significantly improved thermal stability over LiPF₆ at 70 °C. Additionally, a claimed unique feature of LiBOB is its participation in SEI formation through the BOB anion, which allows the use of a pure propylene carbonate (PC)-based electrolyte in graphite electrode-based cells without causing solvent co-intercalation and graphite exfoliation [105]. A new C-2-modified lithium bis(malonato)borate (LiBMB), lithium bis(fluoromalonato)borate (LiBFMB), has been synthesized for LIB applications [111]. The LiBFMB salt has higher oxidation stability than LiBOB, excellent solubility in mixed carbonate solvents, and good cycling performance in 5.0-V LiNi_0.5Mn_1.5O₄-based half-cells.

2.1.2.3 Effect of the Salt Concentration

The salt concentration change in the electrolyte solution in LIBs during operation causes serious degradation of the battery performance. During discharge, the concentration increases at the anode and decreases at the cathode [112], which may in turn influence the microscopic morphologies of the electrodeposited surface [113]. In addition, the local concentration of Li⁺ at the interface between an electrode and an electrolyte solution will increase when forming the SEI at the electrode/electrolyte interface [114, 115].

Bharath et al. studied the effect of concentration from 0.06 to 4 M on transport and structural properties using molecular dynamics simulations [116]. They found that the molecular structure of the solution changes with concentration from a predominantly solvent separated ion pair (SSIP) configuration at the dilute limit to an aggregate-rich configuration at high concentration. This work provides insights into the relation between the molecular structure and performance of electrolyte solutions and suggests ways to design novel electrolytes. It indicates that there is a strong relationship between the surface of the electrode and the concentration of ions in the electrolyte solution. Additionally, Toshiro et al. developed a new method of in situ Raman spectroscopy with ultrafine multifiber probes to study the concentration of ions in the electrolyte solution in deep narrow spaces between the electrodes in batteries [115]. By this method, the permeability, which is a key factor to achieve high battery performance, can be evaluated.

Recently, Shuang-Yan Lang et al. reported the direct visualization (as shown in Fig. 5) of the interfacial evolution and dynamic transformation of sulfides mediated by lithium salts via real-time atomic force microscopy monitoring inside a working battery [93]. The in situ monitoring provides evidence for the mediating effects of Li salts on the interfacial precipitation of Li₂S. The oriented reaction direction of < 200 > dominates the routes and behaviors of Li₂S decomposition, which is strongly correlated with the electrochemical performance of Li – S batteries [93]. The authors reported that the LiTFSI salt induces a 2D growth mode to form lamellar Li₂S during the deposition process, conversely taking edge-to-center and layer-by-layer modes during the decomposition process.

Fig. 5

Copyright © 2018, American Chemical Society

Schematic illustration of the sulfide reactions at the cathode–electrolyte interfaces in the LiTFSI-LiFSI binary salt electrolyte. Reproduced with permission [93].

2.1.3 Nonaqueous Solvents

As one of the most important nonelectrode components, an ideal electrolyte solvent for LIBs needs to meet the following basic requirements: (1) a wide electrochemical window (0–5 V), (2) compatibility with the battery electrode material (no chemical reactions between the electrode and electrolyte), (3) chemical inertness with regard to all cell components during cell operation, (4) high ionic conductivity, low viscosity, and high dielectric constant, and (5) existing as a liquid in a wide temperature range. Its melting point should be as low as possible, while the boiling point should be as high as possible. Any single nonaqueous solvent fulfilling all the requirements at the same time is almost impossible since a high dielectric constant and a low viscosity are usually not integrated into a single compound (with the exception of some organic nitrogen-containing compounds) [117].

The most commonly used solvents in commercial LIBs are organic esters of carbonic acid, among which EC has become indispensable due to a number of properties. Ethers [118], PC [119], EC [120], and linear carbonates [121] are widely used as electrolyte solvents or cosolvents in LIBs.

However, there is a growing interest in replacing these carbonate-based solvents. Kawamura et al. [122] probed the use of methyl difluoroacetate as an electrolyte solvent and reported reduced aluminum corrosion due to the formation of an organic layer on the surface of the current collector.

Fluorination is a commonly used strategy to modify the structure of electrolyte solvents. Kalhoff et al. used two fluorinated derivatives and studied electrolyte solvents in combination with LiTFSI as a conducting salt. The authors concluded that the use of fluorinated linear carbonates as electrolyte (co)solvents is successful in preventing the anodic dissolution of aluminum [123].

2.2 Gel and Solid Electrolytes

Liquid organic electrolytes are the most popular electrolytes in commercial LIBs; however, there are safety issues associated with the high flammability of organic electrolytes that need to be addressed. As the only liquid component in the cell, the electrolyte plays an important role in determining the success of realizing a LIB as a solid device. In electrochemical cells such as batteries and capacitors, ionic pathways provided by electrolytes are required for completing circuits with electrical pathways [105]. Fluidity is not necessarily required since ions can move through the media. Thus, polymer electrolytes, including polymer gel electrolytes, have been actively investigated as a possible replacement for liquid electrolytes. Polymer electrolytes can be divided into solid polymer electrolytes (SPEs), gel polymer electrolytes (GPEs), and quasi-solid electrolytes.

Electrolytes are media that transport ions that are oxidizing or reducing at both electrodes in batteries. Hence, in principle, new battery chemistries require new electrolyte composition. Despite the tremendous progress in cathode-active material development, improvements in electrolytes have not received sufficient research attention and funding in the past [124]. Only recently research has focused on high transference number electrolytes (HTNEs), such as ceramic lithium conductors, SPEs, swollen GPEs, and composite electrolytes in which the anion is less mobile than lithium [125].

2.2.1 GPEs

Polymer gels possess both the strong mechanical properties of solid polymers and the excellent ionic conductivity typical of liquid electrolytes. GPEs can be formed by adding one or more plasticizers to a SPE. GPEs include porous polymer membrane skeletons and liquid organic electrolytes [28]. The performance of GPEs mostly depends on the properties of the polymer host. Polyethylene oxide (PEO), poly(vinylidene fluoride) (PVDF), poly(vinylidene fluoride-hexafluoropropylene) (PVDF-HFP), poly(methyl methacrylate) (PMMA), and polyacrylonitrile (PAN) are employed as polymer hosts for GPEs. Among them, PVDF-based GPEs have been applied to practical applications in lithium batteries [126].

The original concept of SPEs has been extended to GPEs, which work in a manner similar to liquid electrolytes in terms of the electrochemical mechanisms [127]. SPEs remain a laboratory curiosity after nearly four decades. GPEs are much more practical, some of which have already been used in commercial LIBs [105]. Figure 6 shows the difference in the transference number between liquid electrolytes (LEs), GPEs, and SPEs [128].

Fig. 6

Copyright © 2018 WILEY–VCH

Gel electrolytes. a Electrolytes. b Ex situ GPE versus in situ GPE. c Gelation. Reproduced with permission [128].

For example, Baik et al. synthesized a polymerizable lithium salt, lithium (trifluoromethanesulfonyl)(vinylsulfonyl)imide, which was used to prepare cross-linked GPE systems without adding any conventional inorganic lithium salt (Fig. 7). It showed a reasonably high ionic conductivity [129].

Fig. 7

Copyright © 2019, American Chemical Society

Cross-linked polyelectrolyte gels (CPGs) and specific discharge capacity and Coulombic efficiency of the LiFePO₄/CPG₄/Li battery. Reproduced with permission [129].

The advantages of GPEs mainly include the following four aspects: (1) stabilizing Li metal anodes, blocking Li dendrite growth, improving the Coulombic efficiency, optimizing the SEI layer with a selected solvent, and rational design of the anode structure; (2) selection of electrolytes, modifying separators, and enhancing the energy density; (3) increasing the cycling performance under a high current density; and (4) suppressing the polysulfide shuttle effects, preventing large volume expansion, and increasing sulfur utilization for the S cathode (Table 1).

For example, Haiping et al. developed a polymer electrolyte with high conductivity and enhanced mechanical toughness by introducing a double polymer network of PVDF-HFP and PEO into the electrolyte. Benefiting from this newly designed electrolyte, lithium metal anodes showed good cycling performance, retaining a Coulombic efficiency of 96.3% at 400 cycles [130].

A stretchable GPE based on the polyepichlorohydrin terpolymer as a host polymer was developed and studied by Soumyadip et al. The battery showed a 700–800 mAh g⁻¹ initial specific capacity, which dropped slowly with increasing cycle number [131]. To address dendrite growth and build a stable SEI layer, Wei et al. [132] designed a dual-salt (LiTFSI-LiPF₆) GPE with a 3D cross-linked polymer network. As shown in Fig. 8, using an in situ polymerization method, they introduced a dual salt into the 3D structure, which exhibited high ionic conductivity (0.56 mS cm⁻¹ at room temperature) and established a robust and conductive SEI on the lithium metal surface [132].

Fig. 8

Copyright © 2018, WILEY–VCH

Process steps for in situ polymerization of a GPE. Reproduced with permission [132].

A GPE based on renewable polymers such as cellulose triacetate and poly(polyethylene glycol methacrylate) p(PEGMA) was developed at ambient temperature, which is cost-effective and ecofriendly. This GPE exhibited an optimal ionic conductivity of 1.8 × 10⁻³ S cm⁻¹ and a transference number of 0.7 at room temperature. The Li/LiFePO₄ half-cell with the GPE PC exhibited better capacity retention of 83% with a 136 mAh g⁻¹ capacity after 100 cycles [133].

2.2.2 Hybrid Solid-State Electrolytes

Despite the high ionic conductivity and excellent wetting of electrodes, liquid electrolytes suffer from safety concerns such as poor thermal stability, high flammability, and leakage issues. Replacing liquid electrolytes with solid-state electrolytes (SSEs) has been considered the key to safe and durable lithium battery systems using lithium metal anodes. However, inorganic SSEs usually suffer from poor contact with electrodes [134]. Even though significant progress has been made in past years, satisfying the requirements for practical lithium batteries with any single-component SSE seems very challenging. In this context, developing hybrid solid-state electrolytes (HSSEs) has been recently proposed as a feasible approach toward high-performance electrolytes. HSSEs are composed of two or more types of SSEs to compensate for the advantage and disadvantage of each constituent [135]. The most common hybrid electrolyte is a composite electrolyte consisting of a polymer electrolyte and an inorganic SSE. The common solid inorganic electrolytes are listed in Table 2. The most common combination is polymer-inorganic HSSEs. Zhang et al. prepared a PVDF-based composite electrolyte incorporating Li_6.75La₃ Zr_1.75Ta_0.25O₁₂ (LLZTO) ceramics. The ionic conductivity of optimal PVDF/LLZTO composite solid electrolyte membranes can reach a maximum value of 5 × 10⁻⁴ S cm⁻¹ at 25 °C [136].

Table 1 Chemical structures of additives with different functions used for LIBs

Table 2 Summary of inorganic solid electrolytes

A series of interpenetrating polymer network (IPN) SPEs with high stability at high anodic voltages were reported [137]. IPNs incorporate two or more independent networks into one material system. The organic polymers and ionic ceramics of interest have been reported to be inherently incompatible. Their incompatibility results in the formation of a barrier that prevents ion migration across the formed interface. To solve this problem, a thin conformal SiO₂ coating was deposited onto LICGC, followed by salinization with (CH₃CH₂O)₃-Si-(OCH₂CH₂)-OCH₃ in the presence of LiTFSI, which resulted in good adhesion between SiO₂ and LICGC, a low resistance interface, and good wetting of Li⁰ [138].

Li et al. introduced LiF to garnet Li_6.5La₃Zr_1.5Ta_0.5O₁₂ (LLZT) to increase the stability of the garnet electrolyte. Garnet LLZT-2 wt% LiF (LLZT-2LiF) had less Li₂CO₃ on the surface and showed a small interfacial resistance with Li metal, and the all-solid-state Li/polymer/LLZT-2LiF/LiFePO₄ battery displayed a high Coulombic efficiency and a long cycle life [139]. Bi et al. developed a novel hybrid solid electrolyte (HSE) membrane consisting of Li_0.33La_0.557TiO₃ (LLTO) ceramic NRs, PAN, and succinonitrile (SN), which integrated their merits and showed a high ionic conductivity of 2.20 × 10³ S cm⁻¹ at 30 °C. In addition, the HSE was reported to exhibit a high electrochemical window of 5.1 V (vs. Li/Li⁺), superior thermal stability, and good mechanical properties [140].

Yu et al. designed a layered HSE by coating a ceramic lithium aluminum titanium phosphate (LATP) electrolyte with a protective polymer electrolyte, polyphosphazene/PVDF-HFP/LiBOB. The prepared all-solid-state battery with a metallic lithium anode and a high-voltage Li₃V₂(PO₄)₃/CNT cathode showed high capacity and excellent cycling performance with negligible capacity loss over 500 cycles at 50 °C. Furthermore, the analysis of the HSE after long-term cycling demonstrated outstanding electrode/electrolyte interfacial stability [141].

2.3 IL Electrolytes

Electrolytes derived from ILs have gained much attention due to their thermal and electrochemical stability, FR performance [153], and negligible vapor pressure [154]. ILs are salts that are molten below 100 °C, including room temperature. Since they were introduced to polymer-and-salt systems in 1995, ILs have found significant success in improving the overall performance of lithium metal electrodes [155,156,157]. The absence of volatile organic compounds in ILs provides much greater stability at high temperatures compared to the organic electrolytes that are currently widely used in lithium batteries. In addition, ILs possess negligible vapor pressure at room temperature, allowing them to operate in a wide temperature range without performance loss caused by evaporation [158]. Generally, ILs are composed of organic cations and derivatives bonded with inorganic or organic anions (Fig. 9) [159]. The tunable properties of ILs as electrolytes for LIBs have shown great promise enabled by the large number of possible cations and anions, which can eventually yield an unlimited number of combinations for LIBs. Therefore, ILs are considered ideal candidates for substitution of conventional organic electrolytes in LIBs [158, 159]. Furthermore, ILs have been introduced to polymer electrolytes to form hybrid electrolytes by physical blending, chemical functionalization, or polymerization. The resulting hybrid electrolytes showed enhanced mobility caused by the decreasing crystallinity of the polymer. Depending on the combination of polymer and IL electrolytes, IL-based electrolytes can be separated into two categories: organic carbonate and IL electrolytes, and gel polymer and IL electrolytes.

Fig. 9

Copyright © 2020, MDPI

Cation and anion structures of some important ILs. Reproduced with permission [159].

2.3.1 Organic Carbonate and IL Electrolytes

To address the safety problems that conventional organic electrolytes face, ILs have been introduced to enhance the thermal stability of lithium cells. Additionally, ILs possess better conductivity than conventional organic electrolytes. Table 3 lists the common carbonate solvent-based IL electrolytes. Among the different families of ILs, the ILs containing TFSI [(CF₃SO₂)₂ N⁻] have been extensively investigated due to their excellent electrochemical and thermal stabilities [160,161,162,163].

Table 3 Electrochemical properties of IL and carbonate solvent-based liquid electrolytes

To compare the performance of carbonate- versus IL-based electrolytes, Röser et al. studied two selected LiTFSI/pyrrolidinium bis(trifluoromethanesulfonyl imide) room-temperature IL (RTIL)-based electrolytes [164]. Linear sweep voltammetry profiles of the investigated LiTFSI/RTIL electrolytes showed much higher oxidative stability than the state-of-the-art LiPF₆/organic carbonate-based electrolytes at elevated temperatures. The cycling performance of Li-Ni_0.5Mn_1.5O₄ (LNMO)/LTO revealed remarkable improvements with respect to the capacity retention and Coulombic efficiency. In general, the LiTFSI/RTIL-based electrolytes outperform the LiPF₆/organic carbonate-based electrolytes in terms of cycling performance in LNMO/LTO full cells at elevated temperatures.

To further improve the safety and performance of LIBs, rational functionalization of the organic solvent content has been utilized. Ababtain et al. developed novel high-energy–density 3D nanosilica electrodes paired with 1-methyl-1-propylpiperidinium bis(trifluoromethanesulfonyl)imide (Pip) IL/PC/LiTFSI electrolytes [165]. The introduction of PC had no obvious effects on the thermal stability and flammability of the reported electrolytes, while a large improvement in the transport properties was observed. The addition of 20% PC led to a significant improvement in capacity, with stability over 100 charge/discharge cycles.

2.3.2 Gel Polymer and IL Electrolytes

GPEs possess excellent properties, such as high ionic conductivity at ambient temperature, compatibility with both electrodes, a broad electrochemical window, no leakage, and better thermal and appropriate mechanical stabilities, which are desirable for practical application in LIBs [181,182,183]. The introduction of an IL to gel polymers lowers the glass transition temperature of the host polymer and hence improves the ionic conductivity. In addition, ILs can function as a “plasticizer” to reduce the brittleness or increase the flexibility and processability of the polymer electrolyte. Recent literature strongly favors the introduction of ILs into GPEs to greatly enhance the conducting properties and cycling performance. Patel et al. obtained a cross-linked polymer “gel” electrolyte by free radical polymerization of a vinyl monomer in an RTIL electrolyte [N,N-methyl butyl pyrrolidinium bis(trifluoromethanesulfonyl)imide-LiTFSI], which showed a high rate capability in charge/recharge processes compared with the IL [184].

In addition to pristine gel polymer IL electrolytes, nanoparticles have also been applied to enhance the performance of LIBs. Li et al. synthesized a nanoparticle-decorated poly(methylmethacrylate-acrylonitrile-ethyl acrylate) [P(MMA-AN-EA)]-based GPE using the 1-ethyl-3-methylimidazolium bis(trifluoromethanesulfonyl) imide (EMITFSI) IL as a plasticizer [185]. The GPE displayed excellent cycling stability, retaining 95.1% capacity after 100 cycles under a 0.2 C rate in the LiFePO₄/GPE/Li-type cell at room temperature and retained 88.4% of the original discharge capacity at 55  C.

Through in situ polymerization of the 1,4-bis[3-(2-acryloyloxyethyl)imidazolium-1-yl]butane bis[-bis(trifluoromethanesulfonyl)imide] (C1-4TFSI) monomer in an EMITFSI-based electrolyte, which was filled in a poly(diallyldimethylammonium) bis(trifluoromethanesulfonyl)imide (PDDATFSI) porous membrane, Zhou et al. fabricated an HSE, which further advanced the safety of LIBs [186]. The solid electrolyte displayed high ionic conductivity (> 10^–3 S cm^–1 at 25 °C), satisfactory electrochemical stability, inherent incombustibility, and good mechanical strength and flexibility. The in situ assembled LiFePO₄/Li and Na_0.9[Cu_0.22Fe_0.30Mn_0.48] O₂/Na cells using the solid electrolyte of interest exhibited superior cycling performance with high specific capacities. The simple fabrication process and excellent performance of this solid electrolyte pave the way for its application in next-generation LIBs.

3 Polymer Binders

Generally, polymer binders bond the active material and conductive additives on the current collector. Polymer binders ensure the network structure of the active material and conductive additive, thus maintaining the integrity of the electrode. In addition, polymer binders enhance the conductivity and the mass transfer of lithium ions at the interface between the electrodes [187]. Hence, polymer binders play an important role in preserving the mechanical stability, flexibility, and electroconductive pathways of the electrode [188]. In addition to their functions, polymer binders are insoluble in electrolyte solutions, offering excellent chemical and electrochemical properties for LIBs [189].

Notably, there is no polymer binder that can fully satisfy all the requirements discussed above. The most frequently reported polymer binder in the literature is PVDF. It shows excellent electrochemical stability and good capability for absorbing the electrolyte necessary for transporting Li ions to the surface of the active material. However, lithium blocking is observed when PVDF is used as a binder, so searching for alternatives is important. Some natural polymers, such as carboxymethyl cellulose (CMC), alginate, and guar gum, as well as synthetic polymers, such as polyethylene glycol (PEG), PMMA, and poly(vinyl alcohol) (PVA), have been reported in the literature, and they show better performance than PVDF [190]. Su et al. even applied morphology reshaping and reconfiguration concepts to model electrodes, which enabled high-density Li storage through soluble surface/interface organic design and thus fabrication of binder-free electrodes. [191]. CMC is preferred owing to its low cost and environmentally friendly properties. It also improves the ratio of active material in a cell due to the reduction in binder content and faster drying rate during the electrode fabrication process [192].

3.1 Polymer Binders for the Anode Electrode

Silicon is considered one of the most promising candidates for anode materials and has the potential to overcome the theoretical capacity limit of carbonaceous anodes [193]. The gravimetric capacity of Si electrodes is ten times higher than that of state-of-the-art conventional graphite (3 579 mAh g⁻¹ for Si and 372 mAh g⁻¹ for graphite). In addition, Si is an abundant material and has a low working voltage (ca. 0.2 V versus Li/Li⁺) [193, 194]. These advantages make Si the best candidate for next-generation high-capacity anodes [195]. Unfortunately, silicone suffers from a serious volume change (up to 300%) during the lithiation and delithiation processes, which leads to severe particle pulverization and results in an unstable SEI, as well as in gradual deterioration of the electrode capacity [193]. Extensive studies have been conducted to enhance the silicon electrode performance, including designing novel polymer binders [30, 196,197,198]. The commercial PVDF binder was found to not be suitable for Si electrodes due to the weak van der Waals forces between the binder and Si particles [199]. However, the following categories of binders have been shown to be effective for Si anodes: (i) conventional synthetic polymers such as poly(acrylic acid) (PAA) and polyimide; (ii) natural polymers such as CMC and alginate; and (iii) conductive polymer binders [200, 201].

3.1.1 Polymer Binders with a 3D Structure

During the past two decades, many researchers have been developing 3D polymer binders for Si anodes to improve the cycle life performance of LIBs. Rigid binders were found to be incapable of completely preventing the volume expansion of the Si particles and relieving the development of stresses. Additionally, flexible polymers, such as styrene butadiene rubber with high extensibility and low modulus, are not usable as binders for electrodes either [202]. The reason for this is that the polymer chains slide easily with active particles and conductive carbon black additives, which leads to unwanted agglomeration. Therefore, finding or developing a binder with suitable mechanical properties (stiffness/toughness balance) to retain the structural integrity of electrodes is vital.

To address the undesired properties of conventional 3D polymer binders, a novel binder, the poly(acrylic acid sodium)-grafted-CMC (NaPAA-g-CMC) copolymer, was designed by Wei et al. and used for fabrication of stable Si anodes [203]. In contrast to binders with linear (1D) structures, the NaPAA-g-CMC copolymer binder possesses multipoint (3D) interactions with the Si surface, resulting in a stronger binding strength with both Si particles and copper (Cu) current collectors. Additionally, this 3D binder was shown to contribute to the formation of a stable SEI layer on the Si surface. NaPAA-g-CMC-based Si anodes showed improved cycle life performance and higher Coulombic efficiency than those made with well-known linear polymeric binders such as CMC and NaPPA [203]. Meanwhile, a polymerized β-cyclodextrin (β-CDp) binder was also designed and used for the fabrication of Si-based anodes. The β-CDp binder was shown to possess a hyperbranched network structure that offers multidimensional hydrogen-bonding interactions with Si particles and forms stable electrodes. The authors claimed that the β-CDp-based Si electrodes showed improved cycling performance compared to those of other well-known binders, especially when combined with linear polymers at an appropriate ratio to form hybrid binders [204]. Furthermore, a novel binder with a 3D structure was designed using CMC as the backbone and acrylamide (AM) and acrylic acid (AA) as branches [205]. Unlike polymers with linear structures, the recently developed composite binder has a 3D network structure with more contact points for Si particles and contains high-density carboxyl groups, leading to enhanced binding forces between the slurry and the Cu current collectors [206]. Lee et al. also reported an in situ cross-linked CMC-PEG binder used for fabrication of Si anodes and showed a stable cycling performance with a capacity of ∼ 2 000 mAh g⁻¹ up to 350 cycles at a 0.5 C rate [207].

In general, 3D polymeric binders consisting of stiff polymer molecules as backbones and carboxylic acid and/or hydroxylic functional groups linked to the backbone have been proven to be a better binder for Si anodes [208, 209]. Chemical cross-linking of functional binders is regarded as an effective route to establish a 3D interconnected network and therefore increase the stability of the Si anode [210]. The improved cycle life performance based on these polymeric binders represents significant progress in Si anode research.

3.1.2 Water-Soluble Polymer Binders for Si Anodes

A simple and effective copolymer, polyvinyl pyrrolidone/polyaniline (PVP/PANI), was synthesized to serve as a water-soluble conductive binder for Si anodes to address the fast capacity fading and poor cycling performance in LIBs. The PVP/PANI binder combines the good electrical conductivity of PANI and the good resilience of water-soluble PVP. In addition, intermolecular hydrogen bonds and conductive networks are formed between PVP and PANI, ensuring outstanding solubility, chemical stability, and bonding of the copolymer. Si-based anodes fabricated with this binder have demonstrated both enhanced conductivity and stability, which resulted in improved electrochemical performance [211]. In another study, a modified natural polysaccharide (carboxymethylated gellan gum) was investigated as a high-performance water-soluble binder in silicon-based anodes, and improved cycle life performance and decreased capacity fading were reported. Figure 10 shows that the carboxylic and acetylic groups are homogeneously distributed in the modified polysaccharide polymer chain of the binder [212] and thus form strong hydrogen bonds with the surface of the Si particles and the copper current collector. These strong bonds effectively restrict the volume expansion of the silicon particles and help maintain the electronic integrity of the electrodes during repeated charge/discharge cycles. As a result, Si anodes fabricated with the carboxymethylated natural polysaccharide polymer binder possessed high capacity, excellent rate capability, and stable cycle life performance [212].

Fig. 10

Copyright © 2019, American Chemical Society

Chemical structure of polysaccharide and LIB electrochemical performance. Reproduced with permission [212].

3.1.3 Other Polymer Binders for Si Anodes

A conductive MXene (transitional metal carbides, nitrides, and carbonitrides) binder is an example of another type of binder developed for application in high-capacity Si-based anodes. The MXene polymer was synthesized by simple vacuum filtration and used as a multifunctional binder instead of the conventional insulating polymer binders. In the MXene-bonded Si@C film, MXene formed a 3D conductive framework in which Si@C nanocomposites were embedded. This loose and porous structure provided much more space to buffer the large volume expansion of Si@C nanoparticles and thus led to significantly improved cycle life performance compared with conventional CMC- and PVDF-bonded Si@C electrodes. Moreover, the porous structure and the highly conductive MXene offered fast ion transport and very good electrical conductivity for the MXene-bonded Si@C film, which were favorable for its high chargeability rate. These results showed good potential for MXene-bonded Si@C films as anode electrodes for LIBs [213].

A room-temperature cross-linked natural polymer is another binder designed for fabricating stable Si anodes. This binder uses a boronic cross-linker (BC) that spontaneously forms covalent bonds between vicinal alcohols of a commercial polysaccharide polymer (guar gum) and boronic acid groups conjugated to the stiff polystyrene backbone (Fig. 11a). Hydroxylic group-enriched guar gum has good adhesion with the current collector and surface of Si particles and increases the electrode integrity over prolonged cycles. At the same time, the boronic acid-incorporated cross-linker creates a network-type polymer structure through the formation of the covalent bond of two boronic esters without additional driving forces (Fig. 11b) [214].

Fig. 11

Copyright © 2019, WILEY–VCH

Synthesis of BC-g binders. a Schematic of a BC-g binder in a Si electrode showing the formation of a covalent bond from BC, the self-healing ability of abundant hydroxyl groups with PEO and boronic ester, and fast Li-ion transport through PEO groups. b Spontaneous cross-linking reaction of boronic acid and vicinal alcohol at room temperature along with pictures of guar (left) and BC-g binders (right). c Synthesis process of BC. d Fourier transform infrared (FT-IR) spectrum of BC. Reproduced with permission [214].

Synthesis of a binder that is partially carbonized is a good strategy to design a high-performance binder for Si anodes. The partially carbonized polymer acts as both a binder and a conductive additive for the Si anode. A bicomponent Si anode with a partially carbonized binder has been reported to show better electrochemical performance than the traditional Si anode [215].

Graphite is a common anode material widely used in state-of-the-art LIBs. There are a few research reports on polymer binders for graphite electrodes. An allylimidazolium-based poly(ionic liquid), poly[vinylbenzylallylimidazolium bis(trifluoromethane) sulfonylimide] (PVBCAImTFSI), was used as a binder for graphite anodes in Tejkiran’s research work [192]. A reversible discharge capacity of 210 mAh g⁻¹ was obtained for PVBCAImTFSI-based half-cells at a 1 C rate, compared to the value of 140 mAh g⁻¹ obtained for PVDF-based anodic half-cells. After 500 cycles, 95% retention of the discharge capacity was observed. Moreover, PVBCAImTFSI-based anodes exhibited better charge–discharge stability than PVDF-based anodes [192].

4 Separators

Separators, an indispensable part of battery design, are thin, porous membranes that provide electrical insulation between the electrodes while also serving as an electrolyte reservoir for ionic transport [19]. The newly designed separator materials not only act as a robust separator for the electrodes but also facilitate ion transfer and, potentially, stabilize the SEI at both the cathode and anode [216]. The porous structure of the separator is filled with the liquid electrolyte (lithium salt dissolved in a mixture of one or more solvents). In the future, separators will be designed to play a more active role in LIB operation [216].

An ideal separator possesses high mechanical, chemical, electrochemical, and thermal stability. Additional property requirements that determine whether a membrane is suitable as a separator [19] include high wettability for the electrolyte, good permeability, and high porosity for efficient ion transport. The porous properties of the separator need to be tailored to avoid internal short circuits and ensure sufficient infiltration of the electrolyte. The separator must have the proper thickness to achieve a balance between its mechanical properties and Li-ion transport properties [217, 218]. In addition, the separator should be able to stop the diffusion of electrode components or cathode products at elevated temperatures [219]. Moreover, the mechanical properties, ionic conductivity, and tortuosity determine the performance of separators. Some research shows that separators are far from electrochemically passive components [220]. Both the separator structure and the interaction between the pore-space surface and liquid electrolyte impact Li⁺-ion transport and contribute to cell overpotentials [217].

The separators used in LIBs are usually microporous polyolefin membranes, including polyethylene (PE) and polypropylene (PP), because of their good electrochemical stability, suitable mechanical strength, and pore size. PE, PP, and PE/PP separators with pore sizes in the range of micrometers have been commercialized and widely used in LIB technology. However, polyolefin separators normally show low wettability toward the electrolyte and have poor thermal stability, which limits the LIB performance. Therefore, the research interest in this field is concentrated on the modification of polyolefin separators.

4.1 Inorganic/Organic Substance-Modified Separators

To take advantage of SSEs without compromising the properties of the liquid electrolyte, porous solid-state separators with liquid electrolytes were developed [221, 222]. For example, ceramic-coated separators display sufficient mechanical strength and outstanding thermal stability for preventing internal short circuits. Recently, Al₂O₃/PI (copolyimide)-coated PP was employed as a separator in LIBs to improve safety [223], which efficiently reduced the thermal shrinkage (10% at 150 °C) [223]. In distinct contrast, a coating with solely PI was incapable of impeding dimensional changes of the PP separators (22% at 150 °C). The Al₂O₃/PI-coated PP separators achieved rate capabilities and cell performances similar to those of the bare PP separators. Wang et al. developed a novel separator by coating Al₂O₃ particles onto both surfaces of a commercial paper substrate. The as-prepared paper-supported separator showed excellent stability and superior wettability toward the electrolyte and high ionic conductivity compared with conventional PE separators, providing a promising candidate for potential replacement in future LIBs [224]. In addition, anodized aluminum oxide (AAO) membranes with aligned and straight channels meet the requirements of porous solid-state separators. They have been used not only as templates for the synthesis of nanostructured materials [225, 226] but also as separators for LIBs [227, 228]. In Wang et al.’s recent work [230], the AAO separator was reported to lead to a low overpotential during lithium deposition/stripping in lithium symmetric cells. The Li–S cells with AAO separators showed better cycling performance than those with Celgard at high C rates. At a 2 C rate, the cell showed a battery performance of 480 cycles with a capacity degradation rate of 0.105%, which is much better than that when the Celgard separator was used. AAO membranes are rigid and nonflammable, which improves battery safety. The aligned and uniform channels within AAO membranes provide fast pathways for lithium-ion transport, leading to low areal specific resistance.

Jiang et al. reported a strategy to improve the thermal stability and electrolyte affinity of PE separators. By simply grafting the vinylsilane coupling reagent onto the surface of the PE separator using the electron beam irradiation method, followed by a hydrolysis reaction in an Al³⁺ solution, an ultrathin Al₂O₃ layer is grafted onto the surface of the porous polymer microframework without sacrificing the porous structure or increasing the thickness. The grafting method is beneficial for decreasing the interfacial resistance of the separator and increasing the ion transport capability. The as-synthesized Al₂O₃ ceramic-grafted separator (Al₂O₃-CGS) showed almost no shrinkage at 150 °C and decreased the contact angle of the conventional electrolyte compared with the bare PE separator [231].

Mi et al. proposed an electrode-supported thin ceramic film as a separator using a low-cost ceramic material. Thin and porous α-alumina films with a thickness down to 40 nm were coated onto the surface of cathode or anode electrodes by a blade coating method. The organic/inorganic composite separators showed greatly improved physical and chemical stability. The fabrication and cell assembly using this separator are easily scalable. The coated alumina separator has much higher thermal stability and mechanical strength than the PP separator [232].

To improve the stability and wettability of separators, PMMA coated onto PE [233], PVDF coated onto a commercial polyolefin [234], and PVDF-HFP coated onto PE [235] were used as separators. In addition, PEO is another polymer commonly used to improve the wettability of the polyolefin separator [236].

4.2 Modification of Microporous Membrane Separators

Many research papers that reported good performance for LIBs assembled with modified microporous membranes as a separator have been published [237,238,239,240]. A MOF-based ionic sieve (Fig. 12) was designed for the lithium–sulfur battery, functioning as a battery separator to selectively sieve Li⁺ ions while blocking polysulfides [229]. The cooperative combination, in which porous crystalline nanoparticles of the MOF form building blocks while mechanically flexible and robust graphite oxide (GO) laminates compose structural spokes, provides a promising construction strategy for ionic sieve membranes. When a sulfur-containing mesoporous carbon material (approximately 70 wt% sulfur content) was used as a cathode composite without elaborate synthesis or surface modification, the lithium–sulfur battery with a MOF-based separator exhibited a low capacity decay rate (0.019% per cycle over 1 500 cycles). Moreover, there was almost no capacity fading after the initial 100 cycles.

Fig. 12

Copyright © 2016, Nature

Fabrication and structural characteristics of MOF@GO separators. Reproduced with permission [229].

Recently, Sun et al. synthesized zeolitic imidazolate framework-8 (ZIF8) crystals on the surface of cellulose nanofibers (CNFs) and used a ZIF8-CNF membrane as a separator for LIBs. The ZIF8-2-CNF membrane exhibited better surface wettability than a polymer-based separator (13.31° vs. 96.18°). Compared with the LIB fabricated with the three-layer commercial polymer membrane, the LIB fabricated with the ZIF8-CNF membrane had a comparable cycling stability and a better discharge retention stability (88.3% vs. 80.2%) [241].

Mohanta et al. prepared highly porous PAN-based membranes with a higher concentration of LATP ceramic particles by a simple electrospinning technique [242]. PAN with 30 wt% LATP (P-L30) exhibited an enhanced porosity of 90% and was thermally stabler, with the highest electrolyte uptake among all the prepared membranes. LiFePO₄/Li-based coin cells prepared with a P-L30 membrane exhibited an enhanced discharge capacity of 158 mAh g⁻¹ at a 0.5 C rate and displayed good capacity retention with a higher C-rate. Furthermore, only a 13% reduction in the capacity was observed while cycling the coin cell with the P-L30 membrane for 200 cycles at a 0.5 C rate, and an improved Coulombic efficiency of 96.5% was retained after charging and discharging the P-L30-based coin cell for 200 cycles at a 0.5 C rate.

In Sabetzadeh’s work, porous PAN micro/nanofiber membranes were produced by electrospinning a ternary system of PAN/N,N‑dimethylformamide (DMF)/water [243]. The air permeability of the porous PAN micro/nanofiber membrane was higher than that of the nonporous PAN micro/nanofiber membrane and the commercial Celgard PP separator because of the higher porosity (83%). The high porosity of the electrolyte-soaked electrospun nanofiber membrane allowed the ions to migrate more easily, which in turn prevented a temperature rise and improved battery safety. High ion transport through the separator resulted in high ionic conductivity and low electrolyte resistance. Thus, this membrane improved the LIB performance, particularly at higher cycling rates.

A sandwich-structured nano/microfiber-based separator based on nanosized CNFs and microsized glass microfibers (GMFs) was prepared for LMBs [244]. The sandwich-structured separator was composed of 2.5-µm-thick CNF layers coated on each side of a 15-µm-thick intermediate GMF/CNF composite layer. A combination of GMFs and CNFs can meet all the requirements of a good separator. This 20-µm-thick sandwich-structured separator exhibited good thermal stability, excellent electrolyte wettability, low electrolyte resistance, and homogenous surface pore distribution. Due to its macro- and mesoporous structure, the sandwich-structured separator exhibited a high ionic conductivity (1.14 mS cm⁻¹) when soaked with LP40 electrolyte, which increased the rate capability of Li|LFP cells, reaching ~ 110 mAh g⁻¹ at a 5 C rate.

4.3 Environmentally Friendly Separators

The manufacturing process for separators requires high energy input, thus generating a large environmental impact, and for the most part, these components are not recyclable. Recently, there have been attempts to replace these separators with degradable and renewable materials such as biopolymers [245, 246]. Kim et al. synthesized TEMPO (2,2,6,6-tetramethylpiperidine-1-oxyl)-mediated oxidation (TOCN)-based membranes with desired porosity for separators and processed asymmetric mesoporous membranes using a facile papermaking approach. The fabrication process did not contain toxic dispersants or additional pyrogens [247]. These separators made of TOCN with sodium carboxylate surface groups (TOCN – COO – Na⁺) showed significant capacity fading in a few cycles. In contrast, protonated TOCNCOOH membranes showed highly improved electrochemical performance, displaying 94.5% discharge capacity retention after 100 cycles at a 1 C charge/discharge rate.

Moreover, cellulose is also a natural polymer with the widest distribution network and the largest reserves on earth. Compared to synthetic polymers, cellulose has many advantages; for example, it is fully biodegradable, nontoxic, nonpolluting, easily modified, biocompatible, and renewable. Cellulose membranes prepared by the low-cost papermaking process have great potential for replacing conventional polyolefin materials in LIB applications. Mesoporous Cladophora cellulose membranes were reported to be good LIB separators [248]. A nanofibrillated cellulose (NFC) separator is a new type of renewable material composed of a polyethylene terephthalate (PET) layer and a CNF layer. This separator exhibited high porosity (70%), and it could be wetted in a few seconds. It also showed superior electrolyte absorption compared to PP separators. Zhang et al. fabricated thermally resistant northern bleached softwood kraft (NBSK)/polysulfonamide (PSA)/NFC composite membranes for LIBs via a papermaking process [249]. The results showed that the composite membranes exhibit a higher electrolyte uptake (294%) and a higher electrolyte wetting rate than a commercial LIB separator (Celgard 2350).

Li et al. reported rational design and fabrication of a novel type of highly flexible and porous separator based on hydroxyapatite nanowire (HAP NW) networks with excellent thermal stability, fire resistance, and superior electrolyte wettability [250]. The high thermal stability of HAP NW networks enabled the separator to preserve its structural integrity at temperatures as high as 700 °C, and the fire-resistant property of HAP NWs ensured high safety of the battery. Cellulose fibers (CFs) were adopted to enhance the thermomechanical strength of the separator. Electrochemical evaluation indicated that the batteries with the HAP/CF separator exhibited enhanced cyclability and rate capability compared with the commercial PP separator. More encouragingly, the operating temperature of the as-prepared batteries using the HAP NW-based separator and a commercial liquid electrolyte could be increased to 150 °C, benefiting from the high thermal and structural stability of the HAP NW-based separator.

Similarly, a flexible mesoporous redox-active separator composed of a nanocellulose fiber (NCF) and polypyrrole (PPy) composite was fabricated by using a straightforward papermaking process [228] (Fig. 13a). The redox-active separator features a bilayer structure in which one side comprises a (≈ 3 µm) thick insulating NCF layer, while the other side is composed of a 7-µm-thick redox-active PPy/cellulose composite layer. A LIB containing a LiFePO₄ (LFP) cathode and the redox-active separator was shown to exhibit a capacity of 67 µAh cm⁻³ (or 81 mAh g⁻¹) based on the total volume (or weight) of the separator and cathode, which is higher than that (i.e., 18 µAh cm⁻³ or 26 mAh g⁻¹) obtained with a conventional separator. The results showed that the flexible redox-active separator has significant advantages over commercial PE separators in terms of thermal stability and electrolyte wettability.

Fig. 13

Copyright © 2017, WILEY–VCH

Fabrication and morphological characteristics of a redox-active separator. a Schematic illustration of the preparation of the redox-active separator. b Photographs of the flexible redox-active separator. c SEM image of the NCF layer. d SEM image of the PPy-containing layer. e SEM image of a torn redox-active separator. Reproduced with permission [228]

In another work, an anode/separator architecture based on CuO@graphene and CuO-incorporated PVDF-HFP composites was proposed and investigated [251]. The architecture had a robust anode–separator interface and improved the battery performance due to the enhanced interaction between CuO@graphene and the PVDF-HFP/CuO separator, which shortened the Li⁺ diffusion path, accelerated electron transport, and mitigated the volume change of the oxide anode in the electrochemical reactions. Incorporation of 4 wt% CuO into the separator produced a 17% enhancement in the overall capacity of the battery.

Liu et al. fabricated a novel electrospun core–shell microfiber separator with thermally triggered FR properties for LIBs [252]. As shown in Fig. 14, the free-standing separator is composed of microfibers fabricated by electrospinning. During thermal runaway of the LIB, the protective polymer shell melts, triggered by the increased temperature, and the flame retardant is released, thus effectively suppressing the combustion of the highly flammable electrolytes.

Fig. 14

Reproduced with permission from AAAS [252]

Schematic of the “smart” electrospun separator with thermally triggered FR properties for LIBs. a The free-standing separator is composed of microfibers with a core–shell structure, where the flame retardant is the core and the polymer is the shell. The encapsulation of the flame retardant inside the protective polymer shell prevents direct exposure and dissolution of the flame retardant into the electrolyte, preventing the negative effects on the electrochemical performance of the battery. b Upon thermal triggering, the polymer melts, and then, the encapsulated flame retardant is released into the electrolyte, thus effectively suppressing the ignition and burning of the electrolyte.

5 Conclusions and Outlook

In this review, we presented a compilation of the recent developments in each of the major nonelectroconductive components that comprise LIBs. To meet the individual functional needs of the battery, each individual component of today’s LIBs needs to be optimized in future research. While most of the research work has been focused on the development of active materials for advanced anodes and cathodes, the nonelectroconductive components, such as electrolytes, binders, and separators, strongly affect the battery electrochemical performance. By introducing some examples from the literature, we have shown how these components affect the performance of LIBs.

Regarding electrolyte development, research shows that tremendous progress in solid electrolytes and IL electrolytes has been made in recent years. In addition, the importance of additives, solvents, and salts to the electrolytes for LIBs has been discussed. Additives that modify the SEI and improve safety and conductivity have been demonstrated. Current efforts directed at improving the performance and safety of batteries with appropriate additives to electrolytes have been successful, but only a few have been commercialized. In the future, more attention should be paid to designing novel organic compounds with multifunctional groups in a molecule. By using a combination of two or more kinds of additives, the SEI layers on the electrodes could be more flexible and stabler. With continuous improvements in electrolyte additives, we anticipate that they will play a crucial role in the development of next-generation safe and reliable rechargeable LIBs.

Technological advances for electrolytes as well as solvents and salts could play an important role in the development of next-generation electrolytes for LIBs. A proper combination will allow functionality of the battery system. For example, adding different SEI-forming additives will make PC or g-butyrolactone superior solvents. We have summarized some salts and solvents in this paper for researchers to choose the unique one that is most suitable for their system. Moreover, the ideal electrolyte could be a combination of several electrolytes with multiphase composition, such as “liquid + solid-state” or “liquid + solid-state + liquid”. The electrolyte should not only provide adequate solubility and stability for the cathode but also protect the lithium metal, blocking catholyte diffusion and inhibiting dendrite growth. More attention should be paid to the new concepts, such as highly concentrated solutions, high salt-to-solvent ratios, and water-in-salt electrolytes, proposed to design new electrolytes. To further improve the properties of the current electrolytes for LIBs, optimization methods such as varying the constituents of the electrolyte, metal salts, solvents, and additives and their respective ratios should be further investigated. In addition, quasi-solid electrolytes with high ionic conductivity show an opportunity for the development of advanced LIBs. Furthermore, more environmentally friendly water-based electrolytes for LIBs have been a subject of increased research interest.

On the topic of polymer binders, we included the most recent advances reported in the literature, showing various binders for anode and cathode electrodes. The binders can have a significant negative impact on the electrode performance. Whereas most of the polymer binders are PVDF, CMC, and PAA, there are more novel and tailored binders in the LIB field. The world of polymer chemistry is rich with polymers, some of which are even electroconductive, which have yet to be tested as binders for electrode materials.

The main role of the separator is to prevent electrical contact between the positive and negative electrodes of the battery while serving as an electrolyte reservoir to enable ionic transport. The ideal separator should be extremely thin with high mechanical strength, be chemically, electrochemically, and structurally stable, and have a highly porous structure. Currently, modified microporous polyolefin membranes and inorganic substance separators are the most commonly used separators for rechargeable LIBs. Additionally, environmentally friendly separators have the potential to be very important for future separator development. The supramolecular chemistry concept would also be helpful for designing new separators. However, such studies are still in a very early stage.

In summary, it is the authors’ opinion that, for near-term advancement, the most promising areas of research are related to improving the safety of LIBs. The development of nonflammable, nontoxic, and environmentally friendly electrolytes is of primary importance for achieving this goal. The current state-of-the-art carbonate-based electrolytes will be replaced by novel nonflammable electrolytes, IL-based electrolytes, and solid electrolytes. For long-term improvement, the most promising areas for research are related to increasing the specific energy density of the batteries beyond 500 Wh kg^–1. The development of novel high-voltage electrolytes for advanced LIBs is of crucial importance for achieving this long-term goal. Since recently developed cathodes have approached their theoretical capacity, increasing the operating voltage is one of the few remaining options for increasing the specific energy density. All-fluorinated, fluorine-donating, and FSI-structure-like solvent with “full fluorosulfonyl” electrolytes is among the most promising high-voltage liquid electrolytes, which showed stable operation at cell voltages of up to 5.2 V and achieved an energy density of 480 Wh kg^–1. In addition, binders and separators that are stable at higher operating cell voltages have to be developed to achieve long cycle life performance.

Last, we would like to emphasize that a good combination of electrolytes, binders, and separators will have a major effect on the LIB performance. This cannot be ignored during the future development of LIBs. The goal of this review paper is to summarize some of the recent research performed on these nonelectroconductive battery components and, more importantly, to give the battery community newfound insights into this area of research. We hope to inspire future critical thoughts, spark discussion, and promote new design-of-experiments in the LIB research field.