Introduction

Lithium-ion batteries (LIBs) with high energy density, large specific capacity, and long cycle life have been widely applied in the electronics industry. However, due to the anode capacity attenuation and high output potential difference, the safety of LIBs has been generally concerned owing to the battery ignition or even explosion. As a layer of insulation between the cathode and anode, the separator is an important component of LIBs; it works mainly in two ways: to act as an electronic insulator and to provide a lithium-ion transmission channel inside the battery. In general, an ideal separator needs to have good electrical insulation, low ionic resistance, high porosity, operable mechanical strength, electrochemical stability, and excellent electrolyte wettability [1, 2]. Polyolefin separators, which are currently widely used commercially, are based on polypropylene (PP) and polyethylene (PE) [3], although with some advantages in physical and electrochemical aspects, commercial separators are limited by the low ionic conductivity, poor compatibility with electrolytes, and severe shrinkage at high temperature [4,5,6,7], especially in terms of battery safety. This series of questions prompted researchers to reconsider the best candidates for the LIB separator. Under such circumstances, modifications based on PP/PE separators have attracted widespread attention, such as ceramic coating [8,9,10,11,12], electrospinning [13], and preparation of separators.

Gel polymer electrolyte (GPE) combines the safety of solid polymer electrolyte (SPE) with the high ionic conductivity of liquid electrolyte [14], which is considered to be a promising alternative to liquid electrolyte. Polyethylene oxide (PEO) [15, 16], poly(vinylidene fluoride) (PVDF) [17], poly(methyl methacrylate) (PMMA) [18, 19], and poly(vinyl alcohol) (PVA) [20] have been studied as gel polymer matrices. Among these polymers, PVDF is mainly used as a semi-crystalline polymer with relatively good mechanical properties, excellent swelling characteristics, high dielectric constant, and thermal stabilities [17], which has been applied to various applications [21, 22]. However, the dense pore structure of the pristine PVDF membrane restricts the application for the separator in LIBs. In addition, the low ionic conductivity hampered by the high crystallinity of PVDF is also an obstacle. Furthermore, the presence of the strong polar C–F bond deteriorates the compatibility between PVDF and lithium, resulting in the increase in interfacial impedance. Therefore, researchers focus on the preparation of new polymer electrolyte by means of blending different polymers [23,24,25], innovative polymer synthesis [26, 27], and synthesis of composite polymer electrolytes by addition of inorganic fillers [28, 29]. Of the mentioned techniques, blending has been identified as a simple and effective method due to the ease of preparation and property control. Kondawar et al. prepared PVDF/P(VC-VAc) blending polymer electrolytes for lithium-ion battery with LiNi0.5Mn1.5O4 as cathode and showed ionic conductivity at 3.57 mS cm−1 [22]. Zuo et al. developed an ultra-thin PVDF/HEC blending polymer coating layer for modifying th ePE separator with superior cycling stability and ionic conductivity at room temperature [26].

As a hydrophilic substance, PMMA has good compatibility with liquid electrolyte and can improve the affinity between the separator and the electrolyte, thereby improving the electrolyte absorption and ionic conductivity of the separator. A polymer electrolyte composed of PVDF, PMMA, LiClO4, and DMP exhibited the ionic conductivity of about 4.2 mS cm−1 at 30 °C [30]. Yvonne et al. investigated the PVDF/PMMA membranes’ properties with the addition, of CA and the result exhibited the decreased crystallinity due to the addition of CA, the highest porosity of 99.1%, and melting temperature at 162 °C [23]. Numerous studies [31,32,33] have also reported the application of polymer blending in battery separators. Binary blending in the range of blend compatibility can exploit the advantages of polymer materials and may produce new properties. Therefore, a blending membrane based on PVDF/PMMA polymer is prepared to optimize membrane performance.

In the paper, the PVDF/PMMA blending membrane for lithium-ion batteries was developed and fabricated by the non-solvent-induced phase separation (NIPS) method. PVDF/PMMA blending membranes with varied weight ratios were applied to investigate the effect of PMMA on LIB performance, mainly including its compatibility with lithium, morphology, ionic conductivity, thermal stability, and cycle performance.

Experimental

Materials

PVDF (HSV900) was obtained from Arkema Chemical Co., Ltd., and PMMA (5LX-40, Mw = 200,000 g mol−1) was purchased from Longxin Chemical Co., Ltd. (Heilongjiang Province, China). Dimethylacetamide (DMAc) was obtained from Guangdong Guanghua Tech. Co., Ltd., China. The liquid electrolyte was 1.0 M LiPF6 in ethylene carbonate (EC)/dimethyl carbonate (DMC)/ethyl methyl carbonate (EMC) (MEC/MDMC/MEMC: = 1:1:1) (Suzhou Ganmin Chemical Reagent Co., Ltd., China). A commercial PP/PE/PP composite separator (Celgard 2320) was regarded as a contrast with the thickness of 20 μm.

Blending membrane preparation

The poly(vinylidene fluoride) (PVDF)/polymethyl methacrylate (PMMA) blending membrane was prepared by the non-solvent-induced phase separation (NIPS) method. PVDF and PMMA were dissolved into dimethylacetamide (DMAc) with the mass ratio MPVDF/MPMMA = 10:0, 9:1, 8:2, 6:4, respectively, and the total mass percentage of the PVDF/PMMA polymer in the casting solution was 12%. The slurry was stirred in the oil bath at 70 °C for 24 h, then the solution was placed in a vacuum oven for 48 h to eliminate the tiny bubble. After that, the slurry was casted on a glass plate by using a doctor blade and immersed into a coagulation bath (deionized water), then the resulting blending membrane (thickness 70–80 μm) was further immersed into deionized water to remove residual organic solvent for 48 h. Finally, it was air-dried and placed in the vacuum oven. The blending membranes with a mass ratio of PVDF/PMMA = 10:0, 9:1, 8:2, and 6:4 in this paper are denoted as PM-0, PM-1, PM-2, and PM-3, respectively.

Blending compatibility

There are differences among various polymers as for the ability to form a homogeneous system in two or more blends, and the mutual solubility between polymers can be defined by the concept of compatibility [34, 35]. According to Gibbs energy (ΔGm = ΔHm − TΔSm), the mixing process can only proceed spontaneously when the value of the mixed Gibbs free energy is less than zero under normal pressure. The blending components tend to diffuse to each other to achieve a uniform mixture when the compatibility between the blending components is excellent [36].

Based on the concept of material solubility parameter, Bohn [37] proposed the notion of the polymer solubility parameter used to characterize the strength of interaction between simple liquid molecules as in Eq. (1):

$$ \delta ={\left(\varDelta E/{V}_1\right)}^{1/2} $$
(1)

where V1 is the molecular volume of liquid molecule, ∆E is the intermolecular cohesive energy, δ is the material solubility parameter.

Bernard [38] optimized the polymer solubility parameter theory, and there is usually no specific interaction between the blending components, resulting in a low mixing entropy of the blending system [39]; accordingly, the mixed enthalpy can be a substitute for the mixed Gibbs free energy to predict the degree of blend between the various components of the polymer. ΔHm (J cm−3) is calculated by Eq. (2):

$$ \varDelta {H}_{\mathrm{m}}={\left\{{X}_1{M}_1{\rho}_1{\left({\delta}_1-{\delta}_2\right)}^2{\left[\frac{X_2}{X_1{M}_2{\rho}_2+{X}_2{M}_1{\rho}_1}\right]}^2\right\}}^{1/2} $$
(2)

where X1 and X2 are the mass fractions of polymers 1 and 2, respectively (X1 + X2 = 1), M1 and M2 are the relative molecular masses (g mol−1) of the polymer monomer, respectively, ρ1 and ρ2 are the corresponding polymer densities (g cm−3), respectively, and δ1 and δ2 are the solubility parameters of the polymer, respectively.

A large number of experimental data show that [40, 41], when the mixed enthalpy is less than 0.0418 J cm−3, such a binary blend system can be regarded as a compatible system.

Characterization

The surface morphologies of PVDF/PMMA blending membranes were observed through a scanning electron microscope (SEM, JSM5600L, Japan).

The ionic conductivity (σ, mS cm−1) of the membrane was evaluated by an AC impedance spectroscopy of symmetrical battery in order to eliminate the effects of cathode and anode impedance with frequency ranging from 10−2 Hz to 106 Hz over 5 mV of AC amplitude. The ionic conductivity is calculated from Eq. (3):

$$ \sigma =l/\left({R}_{\mathrm{b}}\times A\right) $$
(3)

where Rb is the membrane bulk impedance, and A and l are the effective area and thickness of the membrane, respectively.

The connection between ionic conductivity and temperature can be depicted by Arrhenius equation as shown in Eq. (4):

$$ \sigma ={\delta}_0\exp \left(-{E}_a/ RT\right) $$
(4)

where δ0 means the preexponential factor, Ea is the activation energy, and T and R are the temperature and molar gas constant, respectively.

In addition, the thermal stability of the PVDF/PMMA blending membrane was analyzed through DTA (differential thermal analysis, SDT-Q600) from 100 to 600 °C at a heating rate of 10 °C min−1 in a nitrogen atmosphere. Besides, the dimensional changes were observed after heating at 25 °C, 100 °C, and 160 °C for 30 min. The thermal shrinkage ratio (TSR) was calculated by Eq. (5):

$$ \mathrm{TSR}\ \left(\%\right)=\left({s}_0-s\right)/{s}_0\times 100\% $$
(5)

where s0 and s represent the area of the separator before and after thermal treatment, respectively.

The electrochemical stability was examined by running linear sweep voltammetry (LSV) from 3.0 to 6.0 V at a scan rate of 1 mV s−1. Lithium metal and stainless steel act as counter and working electrodes, respectively. The electrolyte contact angles were measured on a contact angle meter (JGW360, Chengde Chenghui Testing Equipment Co., Ltd., China) by the sessile drop method with 5 μL of electrolyte. Crystal structures of the separators were investigated by X-ray diffraction (XRD, RINT-2000, Japan).

The LiCoO2/electrolyte/Li half-cells were assembled in an argon-filled glove box with oxygen partial pressure less than 1 ppm and sealed in CR2025 coin cells. The LiCoO2 cathode was prepared by blending 80 wt% LiCoO2 (Hunan Shanshan Battery Materials Co., Ltd., China), 10 wt% PVDF as binder (Arkema, Mw = 1,000,000 g mol−1), and 10 wt% carbon black (Super P, Timcal). NMP was added as solvent for wet grinding, and the wet-milled cathode slurry was uniformly coated on an aluminum foil and transferred to a vacuum oven for 24 h, then the dried coated cathode material was cut into a wafer of a specified size by a punching machine and stored in a vacuum oven.

Results and discussions

Miscibility characterization

The mixed enthalpy of the PVDF/PMMA blending system is calculated through Eq. (2) to predict polymer blending. As shown in Fig. 1a, the mixing enthalpy tends to increase gradually and then decrease with the addition of more PMMA, but the value of mixed enthalpy is always lower than 0.0418 J cm−3. It can be seen that the PVDF and PMMA polymers are completely compatible in the blending ratio range from the above discussions. Apart from that, the DTA analyses of glass transition temperatures shown in Fig. 1b also demonstrate the compatibility of polymers. The glass transition temperature (Tg) is the characteristic temperature of amorphous polymer. Tg of pristine PMMA is actually measured to be 102 °C from Fig. 1b while that of PVDF is − 35 °C [42]. So it is difficult to observe the change of Tg in a large interval from Fig. 1b. It should be noted that the shift of the Tg peak is small even under the condition of blending and the DTA curves of PVDF/PMMA blending membranes exhibit no endothermic peak near 102 °C as seen in Fig. 1b, suggesting that there is no crystal phase of PMMA in the PVDF/PMMA blending membrane and the PMMA molecular chain is well dispersed [25].

Fig. 1
figure 1

a Variation of mixing enthalpy with PMMA content in the PVDF/PMMA blending system. b DTA analyses of pristine PMMA and PVDF/PMMA blending membranes

Full size image

Morphological characterization

Figure 2 shows SEM images of the PVDF/PMMA blending membrane which reflect their pore size and distribution. The top surface of the pristine PVDF membrane has a small pore structure with a pore diameter less than 200 nm seen from Fig. 2(A1). As shown in Fig. 2(A1–D1), the pores on the top surfaces of the PVDF/PMMA blending membranes gradually become larger and much more porous as the weight ratio of PMMA increases. So it would appear that the morphology of the PVDF/PMMA blending membrane exhibits the most obvious variety when the PMMA ratio reaches 40% and displays a cross-linked pore structure conducive to the passage of lithium-ions. On the other hand, the bottom surface of the pristine PVDF membrane from Fig. 2(A2) possesses the dense structure and spherical crystals with good crystallinity, which can be ascribed to the slow exchange process of solvent and non-solvent. Contrary to the change in pore size and porosity, the crystallinity of the PVDF/PMMA blending membrane decreases with the increase in PMMA content. It can be seen from the phenomenon that minor pores begin to appear at the bottom of the blending membrane and the spherical crystal disappears. However, the proportion of PMMA cannot be increased indefinitely; PMMA is commonly known as plexiglass with a certain degree of brittleness. Therefore, this study explores that the optimal upper limit of the PMMA mixing ratio is 40%.

Fig. 2
figure 2

SEM image of the top view: A1 PM-0, B1 PM-1, C1 PM-2, and D1 PM-3 membranes, bottom view: A2 PM-0, B2 PM-1, C2 PM-2, and D2 PM-3 membranes

Full size image

X-ray diffraction characterization

The X-ray diffraction pattern helps to provide the effect of PMMA on the structure and crystallization of PVDF in the sample. As shown in Fig. 3, in the case of the pristine PVDF (PM-0) membrane, diffraction peaks at about 18.7° and 26.6° are ascribed to α-PVDF [43]. The presence of a sharp peak at 21.6° and a small peak around 36.8° confirms the formation of the β-phase [44]. It is noted that these peaks have been suppressed when PMMA was added in the PVDF membranes, indicating a significant crystallinity decrease in the PVDF/PMMA blending membranes. Therefore, the PVDF/PMMA blending membranes become more amorphous with higher ionic conductivity.

Fig. 3
figure 3

XRD patterns of PM-0, PM-1, PM-2, and PM-3 membranes

Full size image

Ionic conductivity

Figure 4a shows the temperature-dependent ionic conductivity. It is evident that the rise in temperature contributes to the improvement in electrolyte conductivity from Fig. 4a which may be due to the promotion on thermal movement of ions and polymer chains by the volume expansion of the polymer during heating [45]. Calculated in Eq. (4), the corresponding ionic conductivity of the Celgard 2320 separator and PM-0, PM-1, PM-2, and PM-3 membranes at 26 °C are 0.84, 1.43, 1.55, 2.03, and 2.18 mS cm−1, respectively. It is clearly seen that the ionic conductivity of the PVDF/PMMA polymer electrolyte is significantly superior to that of the Celgard 2320 separator. In terms of bulk impedance, the bulk impedance of the membranes decreases with the increase in temperature as shown in Fig. 4b, in which the bulk impedance of the PM-3 membrane (1.11 Ω) is even smaller than that of the Celgard 2320 separator (1.30 Ω) at room temperature, while the thickness of the PM-3 membrane is more than three times that of the Celgard 2320 separator. Therefore, the PM-3 membrane owns excellent ability to conduct ion on account of the participation of PMMA. The effect of PMMA can be summarized as three main reasons [46]: first, the addition of PMMA has an impact on the porous structure and lithium-ion permeability of the PVDF/PMMA blending membrane. Moreover, the crystallinity of PVDF reduces due to the addition of PMMA. Last but not least, the strong interaction between carbonyl in PMMA and oxygen-containing functional groups in carbonate solvents results in better compatibility between membrane and electrolyte, which strengthens the ion conductivity of the non-crystalline region [46].

Fig. 4
figure 4

a Ion conductivity curves of PM-0, PM-1, PM-2, and PM-3 membranes and the Celgard 2320 separator at different temperatures. b The corresponding bulk impedance change at different temperatures

Full size image

Wettability analysis

The separator with good affinity to the electrolyte can significantly improve the electrochemical performance of the battery. Herein, the wettability of the separator is investigated using contact angle measurements; as shown in Fig. 5, the electrolyte contact angle is 40.21° for the Celgard 2320 separator and 14.53° for the PM-3 membrane. Obviously, the PVDF/PMMA blending membranes have better wettability with the addition of PMMA, which can be ascribed to better compatibility the between membrane and electrolyte resulting from the strong interaction between carbonyl in PMMA and oxygen-containing functional groups in carbonate solvents [46].

Fig. 5
figure 5

Contact angle photographs of the a Celgard 2320 separator and b PM-0, c PM-1, d PM-2, and e PM-3 membranes

Full size image

Thermal stability

Figure 6 compares the dimensional stability of the Celgard 2320 separator with those of the prepared PVDF/PMMA blending membranes at 25 °C, 100 °C, and 160 °C for 30 min. It is worth noting that the shrinkage rates of PM-0, PM-1, PM-2, and PM-3 membranes at 160 °C are 10.3%, 15.2%, 18.1%, and 37.8%, respectively. When the PMMA content reaches 40%, the PM-3 membrane exhibits a reflow phenomenon. Nevertheless, the Celgard 2320 separator shrinks severely at 160 °C owing to the lower melting point of PE materials. Therefore, it can be considered that the PVDF/PMMA blending membrane exceeds the Celgard 2320 separator concerning thermal stability. As presented in Fig. 7, DTA analyses were also performed to investigate thermal stability of PVDF/PMMA blending membranes and the Celgard 2320 separator. The melting endotherm peaks of PM-0, PM-1, PM-2, and PM-3 membranes emerge at 165.0, 164.0, 163.0, and 157.0 °C, respectively. It is obviously observed that the Tm (melting temperature) of the PVDF/PMMA blending membrane decreases with the increase in PMMA content, which means a slight decrease in dimensional stability for the blending membranes. Furthermore, the melting endothermic peak area of blending membranes gradually reduces with more PMMA existing, which indicates fewer crystal phases as well as the decrease in crystallinity of the PVDF/PMMA blending membrane. As for the exothermic peaks corresponding to decomposition of the blending membrane, the initial decomposition temperatures for PM-0, PM-1, PM-2, and PM-3 membranes in Fig. 7 are observed at 456.1, 440.8, 442.2, and 440.5 °C, respectively, while pristine PMMA has an endothermic decomposition peak at 339.6 °C. The result suggests that the addition of PMMA reduces the thermal decomposition temperature of the PVDF/PMMA blending membrane.

Fig. 6
figure 6

Photographs of the Celgard 2320 separator and the PM-0, PM-1, PM-2, and PM-3 membranes before and after heat treatment at 25 °C, 100 °C, and 160 °C for 30 min

Full size image
Fig. 7
figure 7

DTA analyses of pristine PMMA, Celgard 2320 separator, and PM-0, PM-1, PM-2, and PM-3 membranes

Full size image

Electrochemical stability

The electrochemical stability of the PVDF/PMMA blending membrane was characterized by LSV, as shown in Fig. 8, in which forward voltage scanning was performed on Li/electrolyte/SS asymmetric cells at a rate of 1 mV s−1. As the upper limit of electrolyte stability range, the decomposition initiation voltage is generally located at the intersection of the extrapolated linear current in the high-voltage region with the voltage axis [47]. For the Celgard 2320 separator, there is an anodic peak onset at 4.3 V resulting from the decomposition of carbonate solvent. When PMMA is blending with the PVDF polymer matrix, the PVDF/PMMA blending membranes exhibit a higher decomposition voltage up to 4.5 V (PM-0), 4.8 V (PM-1), 5.0 V (PM-2), and 4.7 V (PM-3) than 4.3 V for the Celgard 2320 separator, indicating that the electrochemical stability of prepared separators is suitable for the normal application in LIBs. The enhanced electrochemical stability, on the one hand, may be due to the low crystallinity of PMMA. Besides, the absorption of a large amount of lithium salt solvent in the amorphous phase slows the excessive oxidation of the carbonate solvent [48].

Fig. 8
figure 8

Electrochemical stability window of the Celgard 2320 separator and PM-0, PM-1, PM-2, and PM-3 membranes

Full size image

Compatibility with lithium

In order to investigate the capability of the electrolyte to eliminate polarization in the dynamic cycle, the polarization experiment of the polymer membrane in Li/Li symmetrical cells (charge time/discharge time = 1 h/1 h, current density 0.5 mA cm−2) was performed as illustrated in Fig. 9. Lithium-symmetric batteries do not undergo cyclic short-circuiting when employing PM-0, PM-1, PM-2, and PM-3 membranes, while the Celgard 2320 separator eventually gets short-circuited due to the growth of lithium dendrites. The internal polarization leads to the increase in potential difference as the cycle goes on from Fig. 9. Figure 10a and b show the polarization curves of the interception period 300–310 h in the middle of the cycle and 990–1000 h in the end of the cycle. At 300 h, the voltages centered at 62.5, 45.0, 34.0, 31.0, and 28.0 mV are obviously observed for symmetric cells using the Celgard 2320 separator and PM-0, PM-1, PM-2, and PM-3 membranes, respectively. The lower potential difference plateau means the smaller degree of polarization. Besides, it contributes to the better interface compatibility between the separator and lithium. Concluded from the above discussion, the addition of PMMA improves the compatibility of the separator with lithium. Moreover, at the end of the cycle (990 h), the polarization voltages of symmetric batteries using the Celgard 2320 separator and the PM-0, PM-1, PM-2, and PM-3 membranes turn into 0, 124.0, 86.0, 42.0, and 46.0 mV, respectively, in which the Celgard 2320 separator is short-circuited owing to the piercing of lithium dendrites, resulting in zero voltage. Due to the relatively dense pore structure and low ionic conductivity of the membrane, the polarization inside the cell with the PM-0 membrane gradually aggravates, which leads to the maximum polarization potential difference. In contrast, the PM-1, PM-2, and PM-3 membranes possess excellent ionic conductivity and low interfacial impedance, thus resulting in the minor internal polarization of cells.

Fig. 9
figure 9

Polarization cycle chart of Li/electrolyte/Li cells using the Celgard 2320 separator (a) and PM-0 (b), PM-1 (c), PM-2 (d), and PM-3 membranes (e)

Full size image
Fig. 10
figure 10

Polarization platform curve at 300–310 h (a) and 900–1000 h (b)

Full size image

Cell cycle performance

Figure 11a–e present the 1st, 100th, and 200th charge–discharge profiles of LiCoO2/electrolyte/Li half-cells with different separators, by which the cell polarization condition is reflected. The charge–discharge plateaus of the Celgard 2320 cell becomes inclined with significant difference, suggesting severe polarization and poor cycle stability. However, the voltage difference of cells assembled with PVDF/PMMA blending membranes slightly increases, which implies an efficient reaction process and excellent cycle stability. This conclusion represents better cycling performance for the cells assembled with PVDF/PMMA blending membranes. Moreover, the slow increase in voltage difference between the charge–discharge plateaus of cells employing blending membranes can be considered as normal capacity decay. Apart from that, the cycling performance of the cells based on the Celgard 2320 separator and PVDF/PMMA blending membranes at room temperature under the rate of 1 C can be further compared in Fig. 11f. After 200 cycles, their specific capacities are 83.6, 114.0, 115.1, 128.5, and 130.7 mAh g−1, which belong to the Celgard 2320 separator and the PM-0, PM-1, PM-2, and PM-3 membranes, respectively. It is obvious that the charge–discharge cycling performance of cells with PVDF/PMMA blending membranes is superior to cells with the Celgard 2320 separator. This result can be attributed to the improvement in ionic conductivity and interfacial compatibility. Meanwhile, the columbic efficiency values are close to 100%, which means very low capacitance recession.

Fig. 11
figure 11

Charge–discharge profiles of the cells at 1 C with the a Celgard 2320 separator and b PM-0, c PM-1, d PM-2, e PM-3 membranes, and f cycling stability of the cells at 1 C

Full size image

The rate capabilities of cells based on the Celgard 2320 separator and PVDF/PMMA blending membranes were measured by applying various discharge current densities ranging from 0.2 to 4.0 C every 5 cycles. As seen in Fig. 12, the discharge-specific capacity of cells decreases with the increase in discharge current density. However, among cells with the Celgard 2320 separator and PVDF/PMMA blending membranes, the cell with the PM-3 membrane performs the best discharge-specific capacity at any rate, especially 4 C, at which the specific discharge capacities are 101.8, 105.9, 126.1, 119.6, and 133.3 mAh g−1 for the Celgard separator and the PM-0, PM-1, PM-2, and PM-3 membranes, respectively. Consequently, the above results indicate that the PM-3 membrane possesses excellent electrochemical performances. Therefore, it could be a promising separator for the applications in lithium-ion batteries.

Fig. 12
figure 12

Cycle curves of the Celgard 2320 separator and the PM-0, PM-1, PM-2, and PM-3 membranes

Full size image

Conclusions

This work has presented a gel polymer electrolyte based on PVDF/PMMA blending which are fabricated via the non-solvent phase separation method from the perspective of polymer blending compatibility. The blend of PMMA and PVDF can significantly change the surface morphology of initial membranes, resulting in a dense layer of the PVDF/PMMA blending membrane porous, better compatibility, and enhancement of the ion-conducting ability in the amorphous region. Although the addition of PMMA results in poor performance of PVDF/PMMA blending membranes in terms of thermal stability, it must be noted that PMMA enhances the electrochemical stability of the membrane. Meanwhile, the interface stability between the PVDF/PMMA blending membrane and lithium is also improved. Finally, the specific capacity of the cell assembled with the PM-3 membrane can reach 130.7 mAh g−1 after circulating 200 cycles at the rate of 1 C, which performs better than that with the Celgard 2320 separator. Consequently, the PVDF/PMMA blending membrane can be applied as a promising separator for rechargeable lithium-ion batteries.