Progress in polymeric separators for lithium ion batteries

Abstract

This study reviews the recent developments and the characteristics of polymeric separators used for lithium ion batteries. According to the structure and composition of the separators, they are broadly divided into four types: (1) polyolefin microporous separators, (2) heterochain polymer microporous separators, (3) polymer electrolytes and (4) non-woven separators. In particular, polymer electrolytes were defined as one category of separators for convenient description in this review; these feature intermediates between the two electrodes and possess transport properties comparable with the separator in liquid LIBs. For each category, the structure, characteristics, modification, and performance of separators are described. Finally, guidelines for further improvements in this research are outlined.

---

Hong Zhang

Hong Zhang received her B.S. degree in Chemical Engineering at Fuzhou University in 2011. In the same year, she joined Professor Bao-Ku Zhu's research group at Zhejiang University as a Ph.D. student. Her recent research mainly focuses on the development of lithium ion battery separators from novel polymers.

Ming-Yong Zhou

Ming-Yong Zhou received his B.E. degree in Polymer Science and Engineering from Sichuan University in 2014. In the same year, he joined Professor Bao-Ku Zhu's research group at Zhejiang University as a Master student. His current research focus on the development and implementation of grafted copolymers for energy storage applications.

Chun-Er Lin

Chun-Er Lin is a 3rd year Ph.D. student in the Department of Polymer Science and Engineering at Zhejiang University. His recent research focus on the development of novel polymeric separators for energy storage applications and water treatment.

Bao-Ku Zhu

Bao-Ku Zhu is a professor in the Department of Polymer Science and Engineering at Zhejiang University. He graduated in chemistry at Zhejiang University in 1990. After receiving his Ph.D. degree in polymer chemistry and physics, he spent one year at Chonbuk National University as a post-doctor. His researches include the design and synthesis of functional amphiphilic polymers and the preparation of high performance lithium ion batteries.

---

1. Introduction

The constant increase in global energy demand, together with an awareness of the finite supply of fossil fuels, has sparked the need to explore alternative green energy sources (such as solar, wind and geothermal energy). Besides these factors, increasing pollution due to the growing CO₂ content in the atmosphere has boosted technological efforts to replace the combustion-engine car with electric or hybrid vehicles. However, the success of these technologies relies on the availability of suitable storage systems such as electrochemical batteries.

Among the different varieties of batteries, the lithium ion battery (LIB) is one of the more vigorous competitors due to its high energy and power density, low self-discharge, and long cycle life. LIBs have been already widely used in various portable electronic devices such as cellular phones, laptops, and PDAs. Recently, their applications have expanded to electric and hybrid vehicles. Basically, a LIB consists of four functional components, the anode, the cathode, the separator and the electrolyte, as illustrated in Fig. 1.¹

Fig. 1 Schematic of a typical LIB. This figure was edited from Fig. 2 of ref. 1, with permission from Elsevier.

A separator is placed between the anode and cathode. Its primary function is to prevent physical contact between the electrodes so that free ions rather than electrons can be transported between them.² Although the separator itself is not involved in any cell reactions, its structure and properties considerably influence the interfacial structure and internal resistance of the LIB, thus affecting capacity, polarization, cycle life, and safety. Therefore, the fabrication of a high-performance separator is necessary to develop the overall performance of LIBs.

In principle, an ideal LIB separator must satisfy some basic requirements such as electronic insulation, mechanical and dimensional stability, chemical and electrochemical reliability, excellent wettability in liquid electrolytes, and uniform thickness. Detailed descriptions have been extensively discussed elsewhere and are not included in this review.^3–5 However, it is not easy to achieve all the requirements simultaneously. For example, with the reduction of the separator thickness, the battery's energy and power densities increase, whereas its mechanical strength and safety may deteriorate. Herein, the order of importance of the various requirements varies depending on the application.

2. Types and properties

Based on their structure and composition, we broadly divide polymeric separators into four types: (1) polyolefin microporous separators, (2) heterochain polymer separators, (3) polymer electrolytes, and (4) non-woven separators. In this article, we focus on the influence of the material, processing method, and structure on the properties of the abovementioned separators. In addition, some remarks on the main strengths and weaknesses of each separator as well as future directions are presented. Prior to introducing each system, however, one important concept still has to be clarified. As mentioned earlier, strictly speaking, there are some obvious differences between polymer electrolytes and traditional separators. Polymer electrolytes were defined as one category of separators in this review that features intermediates between the two electrodes and possesses transport properties comparable with the separators in liquid LIBs.^6,7

2.1 Polyolefin microporous separators

Currently, commercial separators are mainly made of polyolefins, including polyethylene (PE) or polypropylene (PP). The dry process and the wet process are the most widely-used methods for preparing the polyolefin separators. In general, separators made using the dry process exhibit distinct slit-pore and straight microstructures, whereas those made by the wet process show interconnected spherical or elliptical pores. Typical surface morphologies of polyolefin separators made by both processes are displayed in Fig. 2.⁴ From the viewpoint of the porous structure, separators made by the dry process are more suitable for high-power-density batteries due to their open and straight-through porous structure, whereas the tortuous and interconnected porous structures of the wet-process separators are more advantageous with respect to suppressing the growth of lithium dendritic structures during long cycle processes. Detailed descriptions of the various fabrication processes as well as the relationships between the structure and properties of polyolefin separators have been reported in numerous monographs and reviews.^8–10

Fig. 2 Surface morphologies of polyolefin separators made by (a) dry process and (b) wet process. This figure was edited from Fig. 3 of ref. 5, with permission from Elsevier.

Although polyolefin separators are basically reliable for portable applications, two major drawbacks have been identified. First, due to their inherent hydrophobicity and low surface energy, the polyolefin separators show poor wettability and retention of liquid electrolytes containing polar solvents such as ethylene carbonate (EC), dimethyl carbonate (DMC), γ-butyrolactone (γ-GBL) and propylene carbonate (PC). Second, the polyolefin separators suffer from severe thermal shrinkage at high temperatures because of their low melting points. Although they have the features of thermal shut-down, which can turn the polyolefin separator into a nonporous insulating film around its melting point, the temperature of the cell often continues to increase even after shutdown before it actually begins to cool.¹¹ Thus, the cell may experience an internal short-circuit or even an explosion, especially when it is exposed to abusive conditions such as overcharging, crushing or overheating. Therefore, a separator with excellent wettability in liquid electrolyte and high thermal stability is highly desired for high performance LIBs.

A multilayer separator made from PE and PP has been patented as a highly heat-resistant separator; this is a perfect combination, providing both safety assurance and sufficient mechanical strength.^12–14 In a multilayered structure (as shown in Fig. 3), the low melting PE layer (135 °C) can act as a thermal fuse and block the pathway of ions when the temperature is close to its melting temperature. At the same time, the higher-melting PP layer (165 °C) retains its dimensional structure and mechanical strength, and thus prevents short circuiting between two electrodes.

Fig. 3 Scanning electron micrographs of the Celgard 2325 (PP/PE/PP) separators: (a) surface and (b) cross-section. This figure was edited from Fig. 5 of ref. 14, with permission from the American Chemical Society.

In addition, many methods have been attempted in the lab such as surface coating, surface grafting and blending.

2.1.1 Surface coating. Surface coating is widely adopted by researchers because of its facile operating process. The coating materials are mainly divided into three categories: functional polar components or gel polymer electrolytes to improve wettability with liquid electrolyte; high temperature polymers such as polyimide and polyamide to improve thermal stability; and inorganic nanoparticles to improve both wettability and thermal stability. The properties of some surface-coated polyolefin microporous separators are listed in Table 1.

Table 1 The properties of surface-coated polyolefin separators

Polyolefin substrate	Coating materials	Liquid electrolyte	Electrolyte uptake (%)	Ionic conductivity (mS cm^−1)	Ref.
PE	AN-MMA	LiClO4–EC/DMC	91	1.00	15
PE	PVDF-HFP/PMMA	LiClO4–EC/DEC	403	1.69	16
PE	PVDF–CAB	LiPF6–EC/EMC	152	2.48	17
PE	PDA	LiPF6–DEC/EC/PC	126	0.41	18
PP	Catechol/polyamine	LiPF6–EC/DMC	270	—	19
PE	PVDF	LiPF6–EC/EMC/DMC	—	0.89	20
PE	Ethylcellulose (EC)	LiPF6–EC/DMC	—	0.68	21
PE	PI (BTDA–TDI/MDI)	LiPF6–EC/DEC	106	0.24	22
PE	SiO2/PMMA	LiPF6–EC/DEC	180	0.74	23
PE	Al2O3 (cPET binder)	LiPF6–EC/DEC	—	—	24
PE	Al2O3 (SBR–CMC binder)	—	82	1.00	25
Polyolefin	PMOTPA	LiPF6–EC/EMC/DMC	—	—	26
PE	SiO2/poly(lithium 4-styrenesulfonate)	LiPF6–EC/DEC	242	0.75	27
PP	SiO2/Al2O3	EC/DEC	120	0.78	28
PE	SiO2/PEI	LiPF6–EC/EMC/DEC	398	0.49	29
PE	Al2O3/PVDF-HFP	LiPF6–EC/DEC	—	0.53	30
PE	DMAET		108	0.29	31
PP	Polyamine/catechol	LiPF6–EC/DMC	270	—	32
PP	Tannic acid	LiPF6–EC/DMC	125	0.46	33

---

Jeong et al.¹⁵ reported a method of reducing interfacial resistance by coating a gellable acrylonitrile (AN)–methyl methacrylate (MMA) copolymer onto the PE separator by dip-coating followed by phase inversion, which can trap a large amount of liquid electrolyte, further facilitating adherence of the electrodes to the separator. Moreover, the cell assembled with this modified separator displayed a stable discharge capacity and an excellent rate performance.

In general, the coating process produces a new interface between the support separator and the coating material, which inevitably compromises ionic conductivity and the electrochemical kinetics of the modified separator. To modify the interfacial properties between the support PE separator and the coated material, Kim et al.³⁴ used gamma-ray irradiation to treat the PE separator before coating the gel polymer electrolyte (PVDF–12% HFP). Compared to the non-irradiated separator, the electrolyte uptake, the ionic conductivity and the cycle performance of the irradiated separator were improved.

Despite success in achieving superior wettability in liquid electrolytes, the all-covered gel coating layer usually causes serious pore blockage and strong resistance to ion transportation, which in turn generates serious negative effects on the ionic conductivity and cycle performances. Park et al.³⁵ introduced closely packed poly(methyl methacrylate) (PMMA) nanoparticle arrays on the surface of the PE separator. In contrast to the conventional dense coating layer, the highly ordered nanoporous structure, i.e. well-connected interstitial voids formed between the closely packed PMMA nanoparticles, not only favored liquid electrolyte wettability but also facilitated ionic conduction.

Inspired by the outstanding adhering ability of mussel, Ryou et al.¹⁸ prepared a polydopamine (PDA)-coated PE separator using a facile and “green” dipping process. The PDA coating decreased the contact angle significantly from 108° ± 1.4° to 39° ± 1.7°, suggesting that the modified PE became more hydrophilic. Afterwards, Wang¹⁹ et al. reduced the manufacturing cost by replacing dopamine with catechol and a polyamine binary system.

Among all the standards that need to be satisfied, the safety issue is directly associated with human life and can thus eventually depress the entire market, which is more critical than others. In fact, the temperature-related safety issue is usually related to the dimensional shrinking or melting of the separator.³⁶ This would result in direct contact and chemical reactions between the strongly oxidative cathode and the reductive anode.

To overcome this drawback, remarkable efforts have been carried out to improve the thermal stability of the polyolefin separators. Chung et al.¹¹ coated a cross-linked polymer synthesized from diethylene glycol dimethacrylate (DEGDMA) on the PE separator. For the uncoated separator, the reduction in the sample size (shrinking), which started at around 120 °C, continued as the temperature increased, and then the sample was completely melted at around 150 °C. On the other hand, the sample coated with polymer synthesized from DEGDMA maintained its original shape until around 150 °C.

In addition, a variety of high heat resistant polymers have also been investigated as coating materials.^21,22 Song et al.²² coated the PE separator with a co-polymerized polyimide (PI), which is composed of 3,3′,4,4′-benzophenone tetracarboxylic dianhydride having 80% toluene diisocyanate and 20% methylene diphenyl diisocyanate. The shrinkage ratio of the PI coated separator was only 10.0% after heating at 140 °C, whereas the bare PE separator showed a significant thermal shrinkage ratio of 83.3% under the same conditions.

Many types of inorganic nanoparticles, such as Al₂O₃,^37,38 SiO₂,^31,39,40 TiO₂ (ref. 41) and BaTiO₃,⁴² were directly coated onto the polyolefin separators to enhance their mechanical strength, wettability and thermal stability. Several properties of the inorganic composite separators, especially ionic conductivity, are influenced by the type and size of the ceramic particle incorporated. Choi et al.⁴³ investigated the effects of SiO₂ particle size on the electrochemical performance of inorganic composite separators. In comparison to 530 nm SiO₂ composite separators, the 40 nm SiO₂ system showed a higher porosity and a shorter tortuous path, which could lead to higher liquid electrolyte uptake and ionic conductivity.

A salient feature of these inorganic nanoparticle-coated composite separators is their unusual porous structure, i.e. the interstitial voids formed between the closely packed nanoparticles are interconnected by a binder. Lee et al.⁴⁴ controlled the morphology of the coating layer by varying the SiO₂/PVDF-HFP ratio. At low SiO₂/PVDF-HFP ratios, SiO₂ domains in PVDF-HFP matrices with nonporous morphologies were developed, whereas PVDF-HFP serves as a binder to interconnect SiO₂ nanoparticles and highly percolated interstitial voids are formed between closely packed SiO₂ nanoparticles at high ratios. In comparison, the composite separator with a high SiO₂/PVDF-HFP ratio showed a higher ionic conductivity and a superior cell performance.

However, the thick inorganic nanoparticle-coated layer inevitably increases the thickness and weight and decreases the ionic conductivity of the separator, although it is helpful in improving the thermal stability.⁴⁵ To reduce the thickness of the inorganic layer, Jung et al.⁴⁶ employed an atomic layer deposition method to introduce inorganic materials into the pores of the PP separator. By this method, the thermal stability of the PP separator is enhanced without any significant change in the thickness.

As the abovementioned inorganic nanoparticles are introduced, the improvement in the separator's properties is mainly related to physical actions without including lithium ion transport processes. Kim et al.^27,47 used core–shell structured SiO₂ particles containing lithium ions in their shells as the functional fillers. The presence of hydrophilic poly(lithium 4-styrenesulfonate) (PLSS) in the shells of the SiO₂ (Li⁺) particles holds the solvent more effectively, which leads to enhancement of interfacial stability. In this coated separator, the liquid electrolyte is not only encapsulated in the pores, but is also retained in the coating layer composed of SiO₂ (Li⁺) particles and P(VDF-co-HFP). As a result, liquid electrolyte uptake is greater than that of the PE separator. As lithium ions dissociated from the SiO₂ (Li⁺) particles can also contribute to the ionic conductivity, the maximum value of the ionic conductivity of the coated separator can be up to 6.0 × 10⁻³ S cm⁻¹ at room temperature.

Potential issues related to inorganic nanoparticle-coated separators are the increased separator weight due to the introduction of heavy inorganic nanoparticles, particle detachment due to insufficient binding, and the unknown electrochemical stability of the inorganic nanoparticles and the binder in the highly oxidizing and reducing environments encountered in LIBs.¹⁰ In addition, during the drying process, care is taken because any drying temperature above 100–120 °C can cause a permanent morphological change in the polyolefin separator skeleton.⁴⁸

Some electroactive redox coating materials, such as poly(4-methoxytriphenylamine) (PMOTPA),²⁶ polytriphenylamine (PTPAn),⁴⁹ and polydiphenylamine (PDPAn),⁵⁰ which can make the microporous separators to effectively shut down during overcharging, have also been coated onto the polyolefin separators. Once the cell is overcharged (i.e. the cathode potential exceeds the oxidation potential of the electroactive polymer), the incorporated polymer will be electrochemically p-doped so as to convert it into a conducting state. This in turn produces a current bypass for the charging reaction through the conductive polymer bridge formed between the two electrodes.

2.1.2 Surface grafting. Many practical problems are encountered during the surface coating process such as uniformity of the coated layer, thickness of the coated layer and adhesion to the substrate. For these reasons, permanent surface grafting, which can immobilize the functional chains, brushes or layers on the separator surface through covalent bonding interactions, is highly desirable for the surface hydrophilic modification of polyolefin separators. In general, graft polymerization can be carried out by UV-irradiation,^51,52 plasma treatment,⁵³ and high energy radiation (γ-rays, ion beams or electron beams).⁵⁴ The uptake and retention of liquid electrolytes were greatly influenced by the degree of grafting and the type of monomers. Table 2 shows the properties of some of the surface-grafted polyolefin microporous separators.

Table 2 The properties of the surface grafted polyolefin separators

Polyolefin materials	Grafting materials	Grafting method	Liquid electrolyte	Electrolyte uptake (%)	Ionic conductivity (mS cm^−1)	Ref.
PE	MMA	Electron beam radiation	LiPF6–EC/DMC	210	1.01	55
PP	PEG	Mussel-inspired dopamine coating	LiPF6–EC/DMC/EMC	145	0.99	56
PE	MMA	γ-Rays	LiClO4–EC/DEC	320	2.0	54
PE	AN	Plasma-induced pre-irradiation	LiPF6–EC/EMC/DEC	—	1.4	53
PP/PE/PP	MMA	Electron beam syn-irradiation technique	LiPF6–DMC/EC	240	1.21	57
PE	MMA	Surface-initiated ATRP	LiPF6–EC/DMC/DEC	195	1.25	58
PE	GMA	Electron beam technology	LiPF6–EC/DMC	—	—	59
PE	Siloxane	Electron beam technology	LiPF6–EC/DMC	—	0.7	60
PP	AA + DEGDM	Electron beam technology	LiAsF6–PC/EC/DME	—	—	61
PE	PVDF-HFP/PEGDMA	Electron beam technology	LiClO4–EC/DEC	125	0.38	62
PE	PMMA	γ-Rays	LiClO4–EC/PC	380	1.3	63

---

Gao et al.⁵⁵ prepared PE-g-MMA separators with different degrees of grafting (DG) using the electron beam radiation technique. The DG was controlled by varying the total electron dose. However, the electrolyte uptake did not increase linearly on increasing DG. The reason can be explained as follows. At a low DG, the separators still retained their pore structures. With an increase in DG, the micropores are gradually filled by PMMA and the porosity decreases.

Li et al.⁵⁷ introduced a pre-irradiation method to avoid the generation of a thick and compact PMMA-grafted layer. In this method, a high-energy radiation source was used to chemically activate the ungrafted polyolefin separators before the graft copolymerization process. Compared with the electron beam-induced syn-irradiation technique, this method produced a much looser and spongier grafted PMMA layer, which could apparently reduce the activation time. Moreover, the grafted separator showed a smooth surface morphology and excellent adhesion properties toward electrodes, which led to a low interfacial resistance.

In a UV-irradiation-assisted grafting process, the introduction of a photo-initiator would inevitably inhibit the electrochemical reactions during battery operation. In a high energy radiation-induced graft polymerization, usual uncontrolled and high energy projectiles, such as electrons, ions, and radicals, often result in the degradation of the polymer backbone (–C–C– bond) as well as the formation of undesirable chemical functionalities and damage to the target molecules. Even worse, these surface grafting approaches usually reduce the separator's mechanical strength. To bypass these disadvantages, Man et al.⁶⁴ developed a hyperthermal hydrogen induced cross-linking (HHIC) technology to graft PEO onto the PP separator. In the HHIC process, energy-controlled hydrogen projectiles collided with both PEO and PP chains to selectively cleave only the –C–H– bonds. The adjacent carbon radicals generated from –C–H– bond cleavages would bond with each other, leading to effective and efficient cross-linking. The PEO layer on the PP separator obtained by HHIC treatment was extremely thin such that the pore size and structure remained intact.

2.1.3 Blending. In surface coating and grafting modifications, the modified sites usually exclude the pores inside the separators due to limited diffusion of the modifying agents into the separator pores. During modification by blending, both the surface and inside-pores of the separator can be modified simultaneously.⁶⁵ However, to date, modification of polyolefin separators by blending has rarely been investigated.

Shi⁶⁵ successfully prepared hydrophilic PE/poly(ethylene-block-ethylene glycol) (PE-b-PEG) blend separators via a thermally-induced phase separation (TIPS) process. It was found that the incorporated polyether chains were greatly enriched in the surface layer and improved the compatibility between separator/liquid electrolyte and the electrodes. Furthermore, such a modified separator matrix could absorb liquid electrolyte and swell to some extent, showing the characteristics of an “active separator” (as shown in Fig. 4).

Fig. 4 Model of the configurations of PE-b-PEG chains in blend separators with different additive contents. This figure was edited from Fig. 3 of ref. 65, with permission from Elsevier.

2.2 Heterochain polymer microporous separators

To develop a high-performance separator with excellent wettability in liquid electrolyte and high heat resistance, various heterochain polymer microporous separators have also been studied such as brominated poly(phenylene oxide) (BPPO),⁶⁶ isotactic poly(4-methyl-1-pentene),⁶⁷ polyoxymethylene,⁶⁸ polystyrene (PS)–PP⁶⁹ blend, and poly(ethylene terephthalate) (PET)–PP⁶⁹ blend polymers.

A series of sponge-like porous polyimide separators have been prepared by Wang et al. using a wet phase inversion process.⁷⁰ The obtained separators exhibited excellent thermal stability and their thermal shrinkage was less than 1% after storage at 200 °C. Moreover, due to their high surface polarity and high porosity, the liquid electrolyte uptake can be up to 190–378%.

A polyetherimide (PEI)-based three dimensional heat resistant skeleton has been constructed via a vapor-induced phase inversion process.⁷¹ This PEI separator can provide good short circuit protection with negligible deformation of the dimensions even at 180 °C due to its relatively high heat deflection temperature (210 °C).

Xiao et al.⁷² prepared pure PVA microporous separators by a non-solvent-induced phase separation (NIPS) process, with notable features such as lower thermal shrinkage, superior electrolyte adsorption/retention capacity and higher ionic conductivity compared to a commercial PP separator.

2.3 Polymer electrolytes

Because of the high oxidation potential of metal oxide cathodes in LIBs, the cell with the abovementioned micropore separator may suffer from decomposition reactions of the liquid electrolyte solutions. Therefore, replacement of the liquid electrolyte with a polymer electrolyte would be highly desirable. The motivations and advantages of using a polymer electrolyte can be listed as follows: (1) suppression of growth of lithium dendritic structures; (2) enhancement of endurance of the varying electrode volume during the cycle process; (3) avoidance of reactivity of liquid electrolytes; (4) improvement in security; and (5) better shape flexibility and manufacturing integrity.^73–75

The development of polymer electrolytes has gone through three stages: (1) solid polymer electrolytes, in which the polymer host itself is used as a solid solvent along with a lithium salt and does not contain any organic liquids; (2) gel polymer electrolytes, in which an organic electrolyte solution is introduced into a polymeric matrix such as PEO,⁷⁶ PVDF,^77–79 and PMMA;^80,81 and (3) composite polymer electrolytes, formed by incorporating inorganic particles into the polymeric matrix.

2.3.1 Solid polymer electrolytes. The study of poly(ethylene oxide) (PEO)-containing lithium salt complexes was first reported by Wright's group in 1973.⁸² Later, Armand et al. highlighted the possible applications of these polymer–salt complexes as polymer electrolytes.⁸³ However, the practical applications of solid polymer electrolytes are hindered by their relatively low ionic conductivity. The ionic conductivity of solid polymer electrolytes, typically less than 10⁻⁴ S cm⁻¹, is restricted by chain mobility above their glass transition temperature.⁸⁴ Many valuable investigations have therefore focused primarily on the enhancement of room-temperature conductivity, via various approaches such as blending,⁸⁵ co-polymerization,^86,87 and cross-linking.⁸⁸ All these improvements have been achieved either by lowering the glass-transition temperature or reducing the crystallinity of the polymers. Some reported polymer electrolytes and their ionic conductivity are listed in Table 3.

Table 3 Some polymer electrolytes and their ionic conductivity

Polymer electrolytes	Lithium salt	Ionic conductivity (S cm^−1)	Ref.
POSS–PEG/methyl cellulose	LiClO4	1.6 × 10^−3	89
HPG–PPEGMA	LiTFSI	1 × 10^−4	90
Oligo(ethyleneoxy)cyclotriphosphazenes	LiN–(SO2CF3)2	1.98 × 10^−4	91
Oligo(ethyleneoxy)cyclotriphosphazenes	LiSO3CF3	1.16 × 10^−4	91
Poly(oxetane)–polyDDOE	LiBF4	3.7 × 10^−5	92
PVA	LiCF3SO3	—	93
P(EO/MEEGE)	LiTFSI	1 × 10^−4	88
PMPEA	LiTFSI	10^−3	94
P(EO/MEEGE)	LiClO4	3.3 × 10^−4	95
P(EO/PO)	LiTFSI	2.2 × 10^−5	96
TEC	LiClO4	1.5 × 10^−4	97
PEP	LiClO4	10^−4	98
P(EO/MEEGE/AGE)	LiTFSI	—	99
PME–GE	LiClO4	10^−4	100
PEO–PEGEEM–MMA	LiClO4	—	101
Siloxy-aluminate polymer	LiAlH4	2.3 × 10^−5	102
PLMA-b-POEM	LiCF3SO3	10^−5	103
Cyanuric chloride/PEO	LiTFSI	3.9 × 10^−5	104
PEG–PDMS	LiTFSI	6.7 × 10^−4	105
PEO	POSS–phenyl-(BF3Li)3	4 × 10^−4	106
TPEO–MA9LC19	LiClO4	2.2 × 10^−5	107

---

To improve the mechanical strength of the solid polymer electrolytes, block copolymer electrolytes, containing both well-defined conducting pathways and a sturdy supporting matrix, have been proposed.¹⁰⁸ This class of polymer electrolytes includes poly(styrene-b-(styrene-g-ethylene oxide)-b-styrene),¹⁰⁹ poly(styrene-b-oligooxyethylene methacrylate-b-styrene),¹¹⁰ poly(styrene trifluoromethanesulfonylimide of lithium-b-ethylene oxide-b-styrene trifluoromethanesulfonylimide of lithium) (P(STFSILi)–PEO–P(STFSILi))¹¹¹ and poly(ethylene glycol)methyl ether methacrylate–polystyrene–poly(ethylene glycol)methyl ether methacrylate (POEM–PS–POEM).¹¹²

Xue⁸⁶ and coworkers reported a branched PI–PEO copolymer and confirmed the possibility of using it as a solid polymer electrolyte for LIBs. Herein, the PEO coil phase allowed high ionic conductivity, whereas the PI rod phase provided good mechanical strength.

Higa et al.⁸⁷ prepared a graft copolymer electrolyte consisting of a PI main chain and POEM side chains using an atom-transfer radical polymerization method. Compared with the solid polymer electrolyte with the PMMA main chain, the tensile strength of this solid polymer electrolyte is about 15 times higher, with an ionic conductivity of approximately 6.5 × 10⁻⁶ S cm⁻¹ at 25 °C.

All the abovementioned solid polymer electrolytes are dual-ion conductors, in which both cation and anion are mobile.¹¹³ The motion of lithium ions is only a small fraction of the overall ionic current, which may lead to a strong concentration gradient and generate internal polarization. To improve the lithium ion transference number of the polymer electrolytes, Bouchet et al.¹¹¹ designed a BAB triblock copolymer P(STFSILi)–PEO–P(STFSILi), in which the anions were immobilized by covalently linking to the polymeric backbone, such that the lithium ion transference number could approach 0.85.

Zhu et al.¹¹⁴ synthesized a novel single-ion conducting polymer electrolyte, lithium polyvinyl alcohol oxalate borate (LiPVAOB), whose monomer is structurally similar to lithium bis(oxalate)borate LiBOB. The obtained polymer electrolyte showed a good electrochemical performance and a stable electrochemical window up to 7 V (vs. Li⁺/Li).

In addition, plastic crystal electrolytes (PCEs), which are prepared by doping lithium salts into a plastic crystalline matrix, have been studied in the laboratory. Due to rotational disorder and the existence of vacancies in the lattice, they usually exhibit fast lithium ion motion.¹¹⁵ For example, a succinonitrile (SN, NC–CH₂–CH₂–CN) plastic crystal electrolyte exhibited a high ionic conductivity of more than 10⁻³ S cm⁻¹ at room temperature.¹¹⁶

2.3.2 Gel polymer electrolytes. Gel polymer electrolytes (GPE) are attractive as they combine the advantages of a liquid electrolyte (high ionic conductivity) and a solid electrolyte (freedom from leaking).^73,117,118 Various types of polymers have been selected as the matrix for GPEs such as poly(ethylene oxide) (PEO), polyacrylonitrile (PAN), poly(methyl methacrylate) (PMMA), poly(vinyl chloride) (PVC), poly(vinylidene fluoride) (PVDF), polyvinyl butyral (PVB)¹¹⁹ and polyvinyl formal (PVFM).¹²⁰

Currently, these available homopolymers cannot simultaneously attain the optimal requirements for mechanical strength, ionic conductivity, and electrochemical stability. To overcome these drawbacks to some extent, copolymers formed by cross-linking or copolymerization with different functional monomers have been used as the matrixes of GPE. These include poly(methyl methacrylate acrylonitrile) (P(MMA-AN)),^121,122 poly(butyl methacrylate styrene) (P(BMA-St)),¹²³ poly(methyl methacrylate acrylonitrile vinyl acetate) (P(MMA-AN-VAc))^124,125 and poly(methyl methacrylate vinyl acetate)-co-poly(ethylene glycol) diacrylate (P(MMA-VAc)–PEGDA).¹²⁶

Costa et al.³ summarized the developments and main characteristics of PVDF and its copolymers in LIB separators. Being a semi-crystalline polymer, PVDF does not easily incorporate liquid electrolyte into the polymer matrix. Moreover, the affinity between PVDF and the liquid electrolyte is not strong enough to hold the entrapped liquid electrolyte. Thus, various PVDF/copolymer blending systems have been prepared to decrease the crystallinity of PVDF as well as to increase its affinity with the liquid electrolyte, and the “active separator” concept was proposed. Herein, the copolymers are mainly amphiphilic copolymers such as P(hexafluorobutyl methacrylate-co-poly(ethylene glycol)methacrylate) (P(HFBMA-co-PEGMA))⁷⁷ and P(methyl methacrylate–poly(ethylene glycol) methacrylate) P(MMA-co-PEGMA).¹²⁷ Table 4 shows PVDF-based gel polymer electrolytes and their properties.

Table 4 PVDF-based gel polymer electrolytes and their properties

Gel polymer electrolytes	Liquid electrolyte	Electrolyte uptake (%)	Ionic conductivity (mS cm^−1)	Ref.
PVDF	LiPF6–EC/EMC/DMC	230	1.3	128
PVDF-HFP	LiPF6–EC/DEC	—	1.2	129
PVDF/PEO-b-PMMA	LiClO4–EC/PC	211	2.79	130
PVDF/PEO–PPO–PEO	LiPF6–DMC/EMC/EC	188	2.94	131
PVDF-HFP/PSx–PEO3	LiTFSI–EC/DMC	520	0.42	132
PVDF–TrFE/PEO	LiPF6–EC/DMC	45	2.3	133
PVDF/P(MMA-co-PEGMA)	LiPF6–EC/EMC/DMC	372	3.01	134
PVDF/PDMS-g-(PPO–PEO)	LiPF6–EC/EMC/DMC	520	4.5	135
PVDF/PMMA	LiPF6–EC/EMC/DMC	129.3	3.38	136
PVDF/PSF	LiPF6–EC/EMC/DMC	129.7	2.03	137
PVDF/P(HFBMA-co-PEGMA)	LiPF6–EC/EMC/DMC	387	3.19	138
PVDF/PVDF–PEO	LiClO4–EC/PC	387	3.03	139

---

To explain the stability mechanism of a liquid electrolyte in the PVDF-based blend, GPE, Li et al.¹²⁷ synthesized a brush-like amphiphilic copolymer P(MMA-co-PEGMA) and blended it into the PVDF matrix. They classified the status of the electrolyte in GPEs into three phases: filled-in pores, swollen in the amorphous phase (swollen phase), and absorbed by the PEG chains that enrich at the separator surface and pore walls (fixed phase), as shown in Fig. 5.

Fig. 5 Status of electrolyte in pure and active separators. This figure was edited from Fig. 11 of ref. 127, with permission from Elsevier.

A dense instead of a porous gel polymer electrolyte for LIBs has been reported by Li et al.¹⁴⁰ Hydroxyethyl cellulose (HEC) has been chosen as the host for this gel polymer electrolyte. The results demonstrated that this dense gel polymer electrolyte can effectively avoid short circuiting in LIBs.

Cai et al.¹⁴¹ used a lithiated perfluorinated sulfonic (PFSA-Li) ion-exchange membrane swollen with non-aqueous solvent as the polymer electrolyte for LIBs. The results demonstrated that a LIB can be operated using only ion-exchange membranes as electrolytes with no doped salts.

2.3.3 Composite polymer electrolytes. The high ionic conductivity of GPE is due to a high liquid electrolyte uptake, but it is accompanied by a loss of mechanical strength and an increase in reactivity toward the lithium metal anode. It is difficult to combine good conductivity and mechanical properties over a wide temperature range. A simple way of improving ionic conductivity as well as mechanical properties is to form a composite polymer electrolyte by adding ceramic fillers. These ceramic fillers are believed to kinetically inhibit crystallization from the amorphous state. X-ray and Raman spectroscopy results show that the effect of fine ceramic fillers can influence ion aggregation in amorphous polyether electrolytes by promoting ceramic–salt interactions.¹⁴²

It has been well established that several parameters influence the electrochemical properties of composite polymer electrolytes: the nature and content of the lithium salt (LiCF₃SO₃, LiClO₄ and LiPF₆), the nature of the nanoparticles (TiO₂, SiO₂ and Al₂O₃), and the size of the nanoparticles. In general, ceramic fillers for the composite polymer electrolytes are broadly classified into two categories: active and passive. Active component materials, such as Li₂N and LiAl₂O₃, are involved in the conduction process, whereas materials such as Al₂O₃, SiO₂ and MgO are passive. Raghavan et al.¹⁴³ prepared a series of nanocomposite polymer electrolytes comprising nanoparticles of BaTiO₃, Al₂O₃ or SiO₂ by an electrospinning technique. Among these particle fillers, BaTiO₃, which has a significant Lewis-acid character, gave the best electrochemical performance. The effect of these ferroelectric ceramic fillers appeared to be highly dependent on the nature of the salt used in the PEO-based systems, with LiClO₄ showing the most dramatic effects.¹⁴⁴ However, in both the cases, the particle size and the characteristics of the ceramic fillers play a vital role in the electrochemical properties of composite polymer electrolytes. The smaller the size of the ceramic fillers, the more they influence the crystallization kinetics of the polymer chains. It has also has been proved that nanosized ceramic fillers showed better compatibility with lithium metal than micron-sized fillers.¹⁴⁵ Table 5 shows the properties of some of the composite polymer electrolytes.¹⁴⁶

Table 5 Properties of some PEO-based composite polymer electrolytes

LiX	Ceramic fillers	Conductivity (S cm^−1)	Ref.
LiCF3SO3	γ-LiAlO2	9.0 × 10^−4	146
LiBF4	γ-LiAlO2	6.0 × 10^−4	146
LiClO4	TiO2	1.8 × 10^−3	146
LiClO4	Al2O3	1.1 × 10^−3	146
LiClO4	BaTiO3	1 × 10^−5	42
LiCF3SO3	ZrO2–TiO2	1.2 × 10^−5	147
LiN(C2F5SO2)2	Nanochitin	—	148
Li(C2F5SO2)2N	BaTiO3	1.5 × 10^−5	149
Li-montmorillonite	Li-montmorillonite	1.6 × 10^−6	150

---

Kumar et al.¹⁴⁵ reviewed state-of-the-art composite polymer electrolytes based on ionic conductivity, lithium ion transference number and electrode–electrolyte interfacial reactions. Stephan et al.¹⁵¹ also reviewed various CPE hosts in view of their electrochemical and physical properties with respect to application in LIBs, mainly focusing on PEO, PAN, PMMA, and PVDF.

2.4 Non-woven separators

Non-woven separators are traditionally prepared by wet processes, such as the paper-making method,^152,153 the solution extrusion method,¹⁵⁴ and the wet-laid method,¹⁵⁵ or by dry processes such as the melt-blown method.¹⁵⁶ In general, these non-woven separators have certain advantages such as relatively low processing costs, high porosity, and low mass. They are not, however, able to compete with polyolefin separators in LIBs due to their inferior tensile strength and excessively large pore size.¹⁵⁵ The large pores would cause internal short circuits, overgrowth of lithium dendrites and self-discharge, and would further reduce the battery cycle performance and safety. Several attempts have made to tackle the abovementioned shortcomings, including the incorporation of inorganic powders,^157,158 fibrillated fibers (poly(paraphenylene terephthalamide) (PPTA)),¹⁵⁵ and gel polymer electrolytes^159,160 into the non-woven separators. Table 6 shows the properties of some non-woven microporous separators.

Table 6 The properties of non-woven microporous separators

Non-woven matrix	Liquid electrolyte	Electrolyte uptake (%)	Ionic conductivity (mS cm^−1)	Ref.
Glass-microfiber/PVA–β-CN	LiTFSI–EC/DMC	—	0.89	161
PP/PVDF	LiPF6–EC/DMC	116	0.65	79
PP/PVDF-HFP/PMMA	LiPF6–EC/DMC	212	1.85	162
PI/PEG	LiPF6	140	1.45	163
PVDF/PEGDA	LiPF6–EC/DMC	—	3.3	164
PET/PVDF-HFP	LiPF6–EC/DEC	—	0.86	2
PET/PVA-co-PE	LiPF6–EC/DMC	160	0.55	165
PE/P(AN-Vac)	LiPF6–DMC/DEC/EC	380	3.8	166
TPU/PVDF	LiClO4–EC/PC	342	3.2	167
PVDF–PVC	LiClO4–PC/EC	290	2.25	168
PVDF–10CTFE	LiPF6–EC/DMC	800	2.57	169
PAN	LiPF6–EC/DEC	—	2.6	170
PET	LiPF6–EC/DMC	484	2.27	171
SiO2/PVDF-HFP coated PET	LiPF6–EC/DEC	1.08 g cm^−3	0.91	43
PMMA coated PET	LiPF6–EC/DEC	160	0.96	172
SiO2 coated PET	LiPF6–EC/DEC	—	0.91	173
PVDF coated PET	LiPF6–EC/DEC	185	0.81	174
PI/PET	LiPF6–EC/DMC/EMC	200	0.66	175
PBT	LiPF6–EC/EMC	—	0.27	176
Cellulose	LiPF6–EC/DEC		0.77	177
Cellulose/PSA	LiPF6–EC/DEC	260	1.2	178
PI–PEO	LiPF6–EC/PC/DMC/EA	178	0.65	159
Carboxymethyl cellulose	LiPF6–EC/DMC	340	1.75	179
PPESK	LiPF6–EC/DMC	1210	3.79	180

---

Degussa has developed a series of Separion (a trade name) separators by coating a porous ceramic layer on each side of a flexible perforated non-woven separator, as illustrated in Fig. 6. In general, the ceramic materials can be alumna, silica, zirconia, or their mixtures and their particle size is required to be a few nanometers. It is proven that the Separion separators have excellent wettability with liquid electrolyte and show negligible shrinkage at high temperatures.

Fig. 6 Schematic of the Separion separators. This figure was edited from Fig. 6 of ref. 5, with permission from Elsevier.

Lee et al.^{44,157,181,182} developed various types of colloidal particles or inorganic nanoparticle-coated non-woven separators. The introduction of poly(methyl methacrylate) (PMMA) colloidal particle arrays on non-woven poly(ethylene terephthalate) (PET) effectively helped to decrease the pore size of the non-woven substrate.¹⁸¹ Moreover, due to the well-connected interstitial voids between the PMMA particles and the strong affinity between PMMA and the liquid electrolyte, the composite non-woven separator allowed facile ion transport and high electrolyte retention. However, it has also been pointed out that the long-term durability of colloidal particle-embedded non-woven separators still needs to be improved.¹⁶³

Zhu et al.¹¹⁸ coated PVDF onto PP/PE non-woven separators. The surface pore size was effectively reduced from 50 mm to 100 nm by the formation of the PVDF gel layer. However, the PVDF gel also permeated into the inner space of the non-woven separator that suppresses the rapid diffusion of lithium ions. To retain large pores in the interior of the non-woven separator, Shi et al.¹⁶³ designed a sandwich-like composite separator, i.e. a crosslinked polyethylene glycol (PEG) skin layer was formed on both sides of the non-woven separator by in situ polymerization. It was demonstrated that the large pores remaining in the interior of non-woven separators could provide quick access for the fast transportation of lithium ions, and thus contribute to a superior rate capability.

Among all the preparation methods of non-woven separators, electrospinning has drawn considerable attention due to its capacity to create a fully interconnected pore structure with a large surface area and high porosity. By controlling the processing parameters, the obtained electrospun separators will be endowed with porosities in the range of 30–90% and pore sizes in the range of sub-micrometers to a few micrometers.¹⁸³

However, the single layer non-woven separator shows unsatisfactory mechanical properties (not free-standing) with respect to its use in LIBs. To address this problem, Angulakshmi et al.¹⁸⁴ prepared a PVDF-HFP/PVC/PVDF-HFP trilayer non-woven separator by electrospinning technology. By adding a second PVC and a third PVDF-HFP layer to the PVDF non-woven separator, the tensile strength of this trilayer non-woven separator became superior to the commercial Celgard separator. Furthermore, due to the increase in fiber diameter, the formation of inter fiber bonding and enhanced crystallinity, heat treatment can be used to improve the mechanical properties of this electrospun non-woven separator.¹⁸⁵

A polymethylpentene (PMP) non-woven separator has been formed directly on the anode by an electrospinning process by Huang.⁴⁸ This PMP separator exhibited much improved thermal stability and interfacial adhesion between the separator and the deposited electrode compared to the commercial polyolefin separator.

Non-woven separators with heat-resistant engineering polymers, such as polyimide (PI),^186–189 polyester (PET),^43,171,190 poly(phthalazinone ether sulfone ketone) (PPESK),^180,191 cellulose,^{152,177,192,193} poly(m-phenylene isophthalamide) (PMIA),^194,195 and aramid,¹⁵³ as the matrix have also attracted lot of attention in recent years. These types of non-woven separators could still retain dimensional stability and effectively isolate the electrodes, further preventing battery short-circuit and explosion at high temperature. They are proven to be promising separators for highly heat-resistant lithium ion batteries.

3. Conclusions and further studies

The separator is a critical component in LIBs and is placed between the cathode and anode to prevent physical contact between the electrodes, so as to isolate the electronic flow while enabling free ionic transport. Although the separator itself is not involved in any cell reactions, its structure and properties considerably affect the internal resistance and cycle performance of LIBs.

Polyolefin separators have been the most dominant ones in commercial lithium ion batteries over the last few decades due to superior properties such as low cost, electrochemical stability, good mechanical strength and pore structure. However, to date, these properties have not all been completely optimized. One of the most serious issues is their hydrophobic nature and lower surface energy, which cause poor compatibility with the liquid electrolyte. Moreover, polyolefin separators suffer from severe thermal shrinkage at high temperatures due to their low melting points. Accordingly, it is important to develop some novel separators for a high-performance and safe LIB.

Polyolefin separators have been modified using various techniques such as surface coating, surface grafting and blending. The modified separators are endowed with certain improved properties such as thermal stability, wettability, and ionic conductivity. Compared to the polyolefin separators, there are two advantages of incorporating inorganic particles. One is suppression of thermal shrinkage and other is enhanced liquid electrolyte wettability. However, the increased thickness and weight inevitably reduces the power and energy densities of the corresponding cell.

Solid polymer electrolytes have several advantages over their liquid counterparts such as no leakage of electrolytes, better shape flexibility, and non-combustible reaction products at the electrode surface that exists in the liquid electrolytes. Furthermore, the incorporation of substantial amounts of organic solvents to form a gel polymer electrolyte has effectively improved the ionic conductivity of polymer electrolytes. However, the requirements of high ion conductivity and good mechanical strength could still not be met simultaneously.

Non-woven separators are now deemed to be important candidates for separators to be used in high power lithium ion batteries due to their low processing cost, high porosity, and low mass. However, to date, non-woven separators are mainly used in rechargeable alkaline batteries. Their application in lithium ion batteries is limited by their low tensile strength and excessively large pore size.

In future, the development of lithium ion battery separators needs to balance performance against the safety and cost. Separators that combine the features of high ionic conductivity and excellent thermal stability are highly desirable for high-performance lithium ion batteries. Moreover, the high cost of separators is mainly due to their production process. Therefore, developing a more cost-effective process is very important with respect to reducing battery separator costs.

References

1. M. Wakihara, Mater. Sci. Eng., R, 2001, 33, 109–134.
2. J. HunáKim, J. Mater. Chem., 2010, 20, 9180–9186.
3. C. M. Costa, M. M. Silva and S. Lanceros-Mendez, RSC Adv., 2013, 3, 11404–11417.
4. H. Lee, et al., Energy Environ. Sci., 2014, 7, 3857–3886.
5. S. S. Zhang, J. Power Sources, 2007, 164, 351–364.
6. A. M. Stephan, Eur. Polym. J., 2006, 42, 21–42.
7. W. H. Meyer, Adv. Mater., 1998, 10, 439–448.
8. M. Weighall, J. Power Sources, 1991, 34, 257–268.
9. G. Venugopal, et al., J. Power Sources, 1999, 77, 34–41.
10. X. Huang, J. Solid State Electrochem., 2010, 15, 649–662.
11. Y. Chung, S. Yoo and C. Kim, Ind. Eng. Chem. Res., 2009, 48, 4346–4351.
12. W.-C. Yu, Shutdown, US Pat., 5691077, 1997.
13. S. E. Hux and W.-C. Yu, US Pat., 5952120, 1999.
14. P. Arora and Z. Zhang, Chem. Rev., 2004, 104, 4419–4462.
15. Y.-B. Jeong and D.-W. Kim, J. Power Sources, 2004, 128, 256–262.
16. J.-Y. Sohn, et al., J. Solid State Electrochem., 2012, 16, 551–556.
17. J. Liu, et al., J. Power Sources, 2013, 226, 101–106.
18. M. H. Ryou, et al., Adv. Mater., 2011, 23, 3066–3070.
19. H. Wang, et al., ACS Appl. Mater. Interfaces, 2014, 6, 5602–5608.
20. L.-F. Fang, et al., J. Appl. Polym. Sci., 2014, 21, 41036.
21. M. Xiong, et al., Carbohydr. Polym., 2014, 101, 1140–1146.
22. J. Song, et al., Electrochim. Acta, 2012, 85, 524–530.
23. J.-H. Park, et al., J. Power Sources, 2010, 195, 8306–8310.
24. Y. Ko, H. Yoo and J. Kim, RSC Adv., 2014, 4, 19229–19233.
25. C. Shi, et al., J. Power Sources, 2014, 270, 547–553.
26. H. Zhang, et al., Electrochim. Acta, 2013, 108, 191–195.
27. W.-K. Shin and D.-W. Kim, J. Power Sources, 2013, 226, 54–60.
28. H. Liu, et al., Ceram. Int., 2014, 40, 14105–14110.
29. Z. Wang, et al., ACS Appl. Mater. Interfaces, 2015, 5, 3314–3322.
30. H.-S. Jeong, et al., J. Power Sources, 2010, 195, 6116–6121.
31. S. M. Kang, et al., Chem. Mater., 2012, 24, 3481–3485.
32. H. Wang, et al., ACS Appl. Mater. Interfaces, 2014, 6, 5602–5608.
33. L. Pan, et al., ACS Appl. Mater. Interfaces, 2015, 29, 16003–16010.
34. K. J. Kim, et al., J. Power Sources, 2012, 198, 298–302.
35. J.-H. Park, et al., J. Power Sources, 2011, 196, 7035–7038.
36. M. Kim and J. H. Park, J. Power Sources, 2012, 212, 22–27.
37. Y. Liao, et al., Electrochim. Acta, 2013, 89, 461–468.
38. T. Lee, et al., Macromol. Res., 2014, 11, 1190–1195.
39. H.-S. Jeong and S.-Y. Lee, J. Power Sources, 2011, 196, 6716–6722.
40. J.-H. Park, et al., J. Power Sources, 2010, 195, 8306–8310.
41. K. M. Kim, et al., J. Electroceram., 2013, 2–3, 146–153.
42. H. Y. Sun, et al., J. Electrochem. Soc., 1999, 146, 1672–1676.
43. E.-S. Choi and S.-Y. Lee, J. Mater. Chem., 2011, 21, 14747–14754.
44. H.-S. Jeong, E.-S. Choi and S.-Y. Lee, Electrochim. Acta, 2012, 86, 317–322.
45. L.-F. Fang, et al., RSC Adv., 2014, 4, 22501–22508.
46. Y. S. Jung, et al., Adv. Energy Mater., 2012, 2, 1022–1027.
47. S. H. Ju, et al., J. Mater. Chem. A, 2013, 1, 395–401.
48. X. Huang, J. Membr. Sci., 2014, 466, 331–337.
49. S. Li, et al., J. Power Sources, 2009, 189, 771–774.
50. S. Li, et al., J. Power Sources, 2008, 184, 553–556.
51. M. Yang, et al., Solid State Ionics, 2005, 176, 2829–2834.
52. H.-S. Kim, et al., J. Power Sources, 2003, 124, 221–224.
53. J. Y. Kim, Y. Lee and D. Y. Lim, Electrochim. Acta, 2009, 54, 3714–3719.
54. S.-J. Gwon, et al., Nucl. Instrum. Methods Phys. Res., Sect. B, 2008, 266, 3387–3391.
55. K. Gao, et al., Electrochim. Acta, 2006, 52, 443–449.
56. L.-F. Fang, et al., J. Membr. Sci., 2013, 448, 143–150.
57. S. Li and K. Gao, Surf. Coat. Technol., 2010, 204, 2822–2828.
58. J.-L. Shi, et al., J. Membr. Sci., 2013, 437, 160–168.
59. J. Ko, et al., Electrochim. Acta, 2004, 50, 367–370.
60. J. Y. Lee, et al., Electrochim. Acta, 2009, 54, 4312–4315.
61. J. L. Gineste, J. Membr. Sci., 1995, 107, 155–164.
62. J.-Y. Sohn, et al., Radiat. Phys. Chem., 2009, 78, 505–508.
63. S.-J. Gwon, et al., Nucl. Instrum. Methods Phys. Res., Sect. B, 2009, 267, 3309–3313.
64. C. Man, et al., J. Mater. Chem. A, 2014, 2, 11980–11986.
65. J.-L. Shi, et al., J. Membr. Sci., 2013, 429, 355–363.
66. J.-J. Woo, et al., Electrochem. Commun., 2013, 35, 68–71.
67. M. B. Johnson and G. L. Wilkes, J. Appl. Polym. Sci., 2002, 83, 2095–2113.
68. S. Wu, et al., Polym. Bull., 2014, 9, 2205–2217.
69. C. Chandavasu, et al., U.S. Pat., 6824680, 2004.
70. H. Wang, et al., Polymer, 2013, 54, 6339–6348.
71. J. Shi, et al., Sci. Rep., 2015, 5, 8255.
72. W. Xiao, et al., J. Membr. Sci., 2015, 487, 221–228.
73. A. Manuel Stephan, Eur. Polym. J., 2006, 42, 21–42.
74. J. Song, Y. Wang and C. Wan, J. Power Sources, 1999, 77, 183–197.
75. F. B. Dias, L. Plomp and J. B. Veldhuis, J. Power Sources, 2000, 88, 169–191.
76. H. Li, et al., Electrochim. Acta, 2011, 56, 2641–2647.
77. H. Zhang, et al., RSC Adv., 2014, 4, 33713–33719.
78. A. Hofmann, M. Schulz and T. Hanemann, Electrochim. Acta, 2013, 89, 823–831.
79. J. Hao, et al., Electrochim. Acta, 2014, 125, 450–456.
80. O. Bohnke, et al., Solid State Ionics, 1993, 66(1), 97–104.
81. H. Zhang, et al., Electrochem. Commun., 2007, 9, 1700–1703.
82. D. Fenton, J. Parker and P. Wright, Polymer, 1973, 14, 589.
83. M. B. Armand, et al., Fast ion transport in solids, ed. P. Vashishta, J. N. Mundy and G. K. Shenoy, North Holland, Amsterdan, 1979.
84. H.-J. Ha, et al., Energy Environ. Sci., 2012, 5, 6491–6499.
85. J. Li and I. M. Khan, Macromolecules, 1993, 26, 4544–4550.
86. C. Xue, et al., Polymer, 2006, 47, 6149–6155.
87. M. Higa, K. Yaguchi and R. Kitani, Electrochim. Acta, 2010, 55, 1380–1384.
88. A. Nishimoto, et al., Macromolecules, 1999, 32, 1541–1548.
89. P. R. Chinnam, H. Zhang and S. L. Wunder, Electrochim. Acta, 2015, 170, 191–201.
90. T. Zheng, et al., Ionics, 2015, 21, 917–925.
91. D. He, et al., Macromolecules, 2012, 45, 7931–7938.
92. Y. Miwa, H. Tsutsumi and T. Oishi, Polym. J., 2001, 33, 927–933.
93. H. Every, et al., Electrochim. Acta, 1998, 43, 1465–1469.
94. M. Kono, E. Hayashi and M. Watanabe, J. Electrochem. Soc., 1998, 145, 1521–1527.
95. M. Watanabe, et al., J. Power sources, 1999, 81, 786–789.
96. M. Watanabe and A. Nishimoto, Solid State Ionics, 1995, 79, 306–312.
97. Y. Ikeda, et al., Electrochim. Acta, 2000, 45, 1167–1174.
98. L. Marchese, et al., Electrochim. Acta, 1992, 37, 1559–1564.
99. S. Matsui, et al., J. Power Sources, 2001, 97, 772–774.
100. M. Kono, et al., Polym. Adv. Technol., 1993, 4, 85–91.
101. J. Gnanaraj, et al., Polymer, 1997, 38, 3709–3712.
102. T. Fujinami, et al., Chem. Mater., 1997, 9, 2236–2239.
103. P. P. Soo, et al., J. Electrochem. Soc., 1999, 146, 32–37.
104. D. M. Tigelaar, et al., Macromolecules, 2006, 39, 120–127.
105. C. N. Walker, et al., ACS Macro Lett., 2012, 1, 737–741.
106. P. R. Chinnam and S. L. Wunder, J. Mater. Chem. A, 2013, 1, 1731–1739.
107. Y. Tong, et al., Appl. Surf. Sci., 2012, 258, 10095–10103.
108. W. S. Young, W. F. Kuan and T. H. Epps, J. Polym. Sci., Part B: Polym. Phys., 2014, 52(1), 1–16.
109. C. Wang, et al., J. Electrochem. Soc., 2003, 150, A1166–A1170.
110. T. Niitani, et al., Electrochem. Solid-State Lett., 2005, 8, A385–A388.
111. R. Bouchet, et al., Nat. Mater., 2013, 12, 452–457.
112. I. M. Khan, et al., Macromol. Chem., 1989, 190, 1069–1078.
113. S. Feng, et al., Electrochim. Acta, 2013, 93, 254–263.
114. Y. Zhu, et al., Electrochim. Acta, 2013, 87, 113–118.
115. D. R. MacFarlane, J. Huang and M. Forsyth, Nature, 1999, 402, 792–794.
116. P.-J. Alarco, et al., Nat. Mater., 2004, 3, 476–481.
117. N. Shubha, et al., J. Power Sources, 2014, 267, 48–57.
118. Y. Zhu, et al., Energy Environ. Sci., 2013, 6, 618–624.
119. F. Lian, et al., J. Membr. Sci., 2014, 456, 42–48.
120. H.-y. Guan, et al., J. Power Sources, 2014, 245, 95–100.
121. M. Rao, et al., J. Membr. Sci., 2008, 322, 314–319.
122. P. Zhang, et al., Electrochem. Commun., 2008, 10, 1052–1055.
123. Y. Liao, et al., J. Membr. Sci., 2010, 352, 95–99.
124. Y. Liao, et al., Electrochim. Acta, 2009, 54, 6396–6402.
125. Y. Liao, et al., J. Power Sources, 2009, 189, 139–144.
126. Y. Liao, et al., in Meeting Abstracts, The Electrochemical Society, 2010, vol. 1, p. 170.
127. H. Li, et al., Electrochim. Acta, 2014, 115, 317–325.
128. G.-L. Ji, et al., Polymer, 2007, 48, 6415–6425.
129. W. Pu, et al., J. Membr. Sci., 2006, 272, 11–14.
130. Q. Xiao, et al., J. Membr. Sci., 2009, 334, 117–122.
131. Z.-Y. Cui, et al., J. Membr. Sci., 2008, 325, 957–963.
132. S. Seidel, et al., Chem. Commun., 2015, 51, 12048–12051.
133. C. Costa, et al., J. Power Sources, 2014, 245, 779–786.
134. H. Li, et al., Electrochim. Acta, 2014, 115, 317–325.
135. H. Li, et al., Electrochim. Acta, 2014, 116, 413–420.
136. T. Ma, et al., J. Membr. Sci., 2013, 444, 213–222.
137. Q. Cheng, et al., J. Power Sources, 2014, 266, 401–413.
138. H. Zhang, et al., RSC Adv., 2014, 4, 33713–33719.
139. F. Deng, et al., J. Membr. Sci., 2015, 491, 82–89.
140. M. Li, et al., J. Membr. Sci., 2015, 476, 112–118.
141. Z. Cai, et al., Energy Environ. Sci., 2012, 5, 5690–5693.
142. A. Best, et al., Solid State Ionics, 1999, 126, 269–276.
143. P. Raghavan, et al., Electrochim. Acta, 2008, 54, 228–234.
144. M. Forsyth, et al., Solid State Ionics, 2002, 147, 203–211.
145. B. Kumar and L. G. Scanlon, J. Power Sources, 1994, 52, 261–268.
146. F. Croce, et al., Electrochim. Acta, 2001, 46, 2457–2461.
147. L. TianKhoon, et al., Solid State Ionics, 2015, 276, 72–79.
148. N. Angulakshmi, et al., Electrochim. Acta, 2010, 55, 1401–1406.
149. Q. Li, et al., Ionics, 2002, 8, 79–84.
150. R. A. Vaia, et al., Adv. Mater., 1995, 7, 154–156.
151. A. Manuel Stephan and K. S. Nahm, Polymer, 2006, 47, 5952–5964.
152. Q. Xu, et al., RSC Adv., 2014, 4, 7845–7850.
153. J. Zhang, et al., Solid State Ionics, 2013, 245–246, 49–55.
154. S. J. Law, H. Street and G. J. Askew, US Pat., 6358461, 2002.
155. W. Yi, et al., J. Power Sources, 2009, 189, 616–619.
156. K. Singha, et al., Int. J. Textil. Sci., 2012, 1(1), 7–14.
157. J.-R. Lee, et al., J. Power Sources, 2012, 216, 42–47.
158. X. Huang, D. Bahroloomi and X. Xiao, J. Solid State Electrochem., 2014, 18, 133–139.
159. J. Shi, et al., J. Mater. Chem. A, 2014, 2, 9134–9141.
160. Y. Zhu, et al., Energy Environ. Sci., 2013, 6, 618–624.
161. Q. Wang, et al., Electrochim. Acta, 2015, 157, 191–198.
162. D. Wu, et al., J. Power Sources, 2015, 290, 53–60.
163. J. Shi, et al., J. Power Sources, 2014, 271, 134–142.
164. K. M. Kim, et al., Electrochim. Acta, 2014, 120, 159–166.
165. M. Xia, et al., J. Power Sources, 2014, 266, 29–35.
166. X. Li, et al., J. Appl. Electrochem., 2010, 40, 2185–2191.
167. N. Wu, et al., J. Power Sources, 2011, 196, 8638–8643.
168. Z. Zhong, et al., Mater. Sci. Eng., B, 2012, 177(1), 86–91.
169. F. Croce, et al., Energy Environ. Sci., 2011, 4, 921–927.
170. T.-H. Cho, et al., J. Power Sources, 2008, 181, 155–160.
171. J. Hao, et al., J. Membr. Sci., 2013, 428, 11–16.
172. J. HunáKim, J. Mater. Chem., 2011, 21, 8192–8198.
173. H.-S. Jeong, et al., J. Membr. Sci., 2012, 415, 513–519.
174. H.-S. Jeong, et al., Electrochim. Acta, 2011, 56, 5201–5204.
175. J. Ding, Y. Kong and J. Yang, J. Electrochem. Soc., 2012, 159, A1198–A1202.
176. C. J. Orendorff, et al., Adv. Energy Mater., 2013, 3, 314–320.
177. S.-J. Chun, et al., J. Mater. Chem., 2012, 22, 16618–16626.
178. Q. Xu, et al., ACS Sustainable Chem. Eng., 2013, 2, 194–199.
179. J. Zhang, et al., ACS Appl. Mater. Interfaces, 2012, 5, 128–134.
180. W. Qi, et al., Mater. Lett., 2012, 66, 239–241.
181. J.-H. Cho, et al., J. Mater. Chem., 2011, 21, 8192–8198.
182. H.-S. Jeong, et al., J. Membr. Sci., 2012, 415–416, 513–519.
183. J. R. Kim, et al., Electrochim. Acta, 2004, 50, 69–75.
184. N. Angulakshmi and A. M. Stephan, Electrochim. Acta, 2014, 127, 167–172.
185. Y. Liang, et al., J. Power Sources, 2013, 240, 204–211.
186. J. Lee, et al., J. Power Sources, 2014, 248, 1211–1217.
187. Y.-E. Miao, et al., J. Power Sources, 2013, 226, 82–86.
188. Q. Wang, et al., Electrochim. Acta, 2014, 132, 538–544.
189. J. Lee, et al., J. Power Sources, 2014, 248, 1211–1217.
190. M. Xia, et al., J. Power Sources, 2014, 266, 29–35.
191. C. Lu, et al., J. Appl. Electrochem., 2013, 43, 711–720.
192. J. Zhang, et al., Sci. Rep., 2014, 4, 3939.
193. G. Ding, et al., J. Electrochem. Soc., 2015, 162, A834–A838.
194. Y. Zhai, et al., J. Mater. Chem. A, 2014, 2, 14511–14518.
195. K. S. Jeon, et al., Mater. Lett., 2014, 132, 384–388.

---