2.1. Materials

PE separator was donated from WIDE Company, Korea. Acrylamide (AAm, purity>99%) was purchased from J&K Chemical Ltd. Liquid electrolyte, 1 M LiPF₆ in ethylene carbonate/dimethyl carbonate (EC/DMC, 1:1 v/v) was supplied by Beijing Institute of Chemical Reagents and stored in Ar atmosphere. All reagents were used without purification.

2.2. Surface modification of PE separators

We prepared grafted PE separator by co-irradiation induced graft polymerization technique. Strips of PE separators were soaked in acetone to clean the surface, vacuum-dried at 60℃, and weighed. AAm solution were prepared by adding the AAm monomer and 0.5wt% CuSO₄·5H₂O into methanol under stirring. The CuSO₄·5H₂O was used as inhibitor in the reaction. Then, the weighed PE strips were immersed into the solution. Under the nitrogen gas protection, the mixture was irradiated by γ-ray in a ⁶⁰Co source at a dose rate of 0.49 Gy/sec for 17 h. The grafted separators were washed thoroughly and soaked overnight with hot distilled water to ensure extraction of residual monomers and homopolymers. After that, the strips were dried at 60℃ for 24 h and weighed.

2.3. Characterization of separators

The degree of grafting (DG) is defined as

where W₀ and W₁ are the weight of PE and PE-g-PAAm separators, respectively.

Structural changes of the separators were evaluated by Fourier transform infrared (FT-IR) spectroscopy on a Nicolet Avatar 370 FT-IR spectrometer by the culmination of 32 scans, at a resolution of 4 cm⁻¹, and X-ray photoelectron spectroscopy (XPS) on a Kratos Axis Ultra instrument, with monochromatic Al Kα ray, using wide scans in the range of 1100–0 eV, and narrow scans for the C 1s region. Scanning electron microscope (SEM) images of the separators were taken on a Hitachi SEM (S-4800, Japan) after sputtering gold on the sample. The voltage was 25 kV and the current was 10 mA. Different scanning calorimetry (DSC) test was conducted with a DSC Q100 system of TA instrument under nitrogen atmosphere at a heating rate of 10℃/min. The degree of crystallinity (X_c) was calculated by Eq. (2):

where ΔH_m is the heat of melting sample obtained from the peak area and ΔH₁₀₀ is the heat of melting of 100% crystalline PE polymer, which is 288.0 kJ/kg[19].

Liquid electrolyte uptake (ΔW) of separators was calculated by weight changes of separators. We used the tissue to remove the excess liquid electrolyte sticking to the surface of separator quickly before weighing.

where W_a and W_b are the weight of samples before and after the 24-h soaking at room temperature, respectively.

The separators were immersed in the liquid electrolyte, and then clipped by two platinum electrodes for measuring ionic conductivity (σ), which was calculated by Eq.(4). The impedance (R_b) was obtained by measuring AC impedance analysis from electrochemical work station system (CHI66d) at 30℃–70℃ over a frequency range of 1 Hz–10⁵ Hz at 5 mV.

where d and A are the thickness and effective area of the separator, respectively.

To investigate the cycle performance, the coin cells assembled as LiCoO₂ (cathode)/separator-liquid electrolyte/lithium (anode) were tested on the cycle instrument (LAND, China) at room temperature, in the voltage range of 2.8–4.15V. The specific capacity (D) was calculated by Eq.(5):

where C is capacity of the tested battery and M is weight of the LiCoO₂ on the cathode sheet.

3.1. Graft polymerization study

As shown in Fig.1, the DG increased with the AAm concentration. Before AAm concentration of 6.2 wt%, DG increased to 5.1% slowly, as there was only a small quality of PAAm in the system. Then, the DG increased rapidly, reaching 33.6% at AAm concentration of 14.2wt%, where the DG increase became slowly again, towards a level of saturation. These can be explained as follows: with two kinds of radical in the system, there were competition between grafting reaction and copolymerization, when the AAm concentration was below 14.2wt%, the grafting prevailed; while the copolymerization grew stronger with more AAm, the system became viscous gradually, which further hampered the grafting [20].

Fig. 1.Full-screen View

Effects of mass fraction of AAm on DG of PE-g-PAAm separators

3.2. FT-IR and XPS characterizations

Surface chemical compositions of the PE and PE-g-PAAm separators were analyzed by FT-IR and XPS spectroscopy. The FT-IR spectra of the pristine PE and PE-g-PAAm separators of various DGs are shown in Fig.2(a). After grafting, new peaks at 3417 and 3347 cm⁻¹ were identified as the asymmetric and symmetric stretching vibration bands of the N–H group, and the new bands at 3197 and 1712 cm⁻¹ stood for C(O)-NH₂ and C=O stretching vibration of grafted PAAm. The peak intensities increased with DG, indicating that PAAm chains were grafted onto the surface successfully [15].

Fig. 2.Full-screen View

FT-IR (a) and XPS (b) spectra of the pristine PE and PE-g-PAAm separators of different DGs

Fig.2(b) shows the C1s wide-scan XPS spectra for the pristine PE and PE-g-PAAm separator (DG 22.6%).There were only C and O in pristine PE, while N appeared in PE-g- PAAm separator, an indication of PAAm graft onto the PE separator. The tiny amount of O in pristine PE separator may be the surfactant that was not informed by the supplier.

The XPS core-level spectra for the pristine PE and PE-g-PAAm separator (DG 22.6%) are shown in Fig.3. The PE spectra had just one peak at 284.8 eV, which was the overlap of CH₂ and CH₂-CH₂ groups. In the PE-g-PAAm spectra, two new peaks at 286.8 and 288.2 eV were attributed to the amide groups [15]. The FT-IR and XPS results indicate that PAAm graft chains are combined to the PE chain by chemical bonds.

Fig. 3Full-screen View

C1s spectra of (a) pristine PE separator (b)PE-g-PAAm separator with DG of 22.6%

3.3. Morphologies of the separators

SEM images (Fig.4) show that both the pristine separator and grafted separators of different DGs had many open pores distributed uniformly. The PAAm-grafted separators retaining their microporous net structure, but the microporous structure of PE-g-PAAm45.9% became worse apparently.

Fig. 4Full-screen View

SEM images of pristine PE separator (a) and PE-g-PAAm separators of DG being 16.7% (b), 33.6% (c) and 45.9% (d).

3.4. Thermal properties

Thermal properties of PE and PE-g-PAAm separators of different DGs measured by DSC are given in Fig.5 and Table 1. The thermograms of pristine and grafted separators remained almost homologous, but the intensity of peaks decreased with increasing DGs. Also, the onset, peak and degree of crystallinity decreased with increasing DGs. The reduction in the degree of crystallinity upon grafting of PAAm corresponded to the dilution of PE crystalline zone with the amorphous PAAm and the partial disruption in the crystalline structure of PE separator.

Fig. 5Full-screen View

DSC curves of PE and PE-g-PAAm separators with different DGs

3.5. Liquid electrolyte uptake

The affinity to electrolyte is a key property of a separator, and higher uptake means more electrolyte contained in separator. The uptake of electrolyte is shown in Fig.6. As we know, the porosity and the content of active component of separator are the factors that determine the electrolyte uptake.[1]. The uptake increased with the DG and rose to 688.4% at DG=22.6%, where it began to decrease. This maximum increase is almost 2.8 times of the pristine PE(245.0%), because the polar functional groups of the grafted PAAm can greatly improve the affinity between the separator and electrolyte, hence the increase of electrolyte uptake [21]. When DG was greater than 22.6%, too many separator pores were occupied by the grafted chains, which hindered the swelling of the separators and resulted in the decreasing uptake.

Fig. 6Full-screen View

Liquid electrolyte uptake of PE-g-PAAm separators of different DGs.

3.6. The lithium ion conductivity of the separators

Lithium ion conductivities of the pristine PE separator and PE-g-PAAm separators of different DGs, as function of temperature, are shown in Fig.7. Generally, ionic conductivity is influenced by the active component and the mobility of lithium ions [22]. Uncovered open pores, good connectivity of separators, and high amount of lithium ions, which depends on the high electrolyte uptake, can make sense of lithium ion mobility and improve ion conductivity^[1]. At DGs≤ 22.6%, the electrolyte contained in the separators, and their conductivity, increased with DG, as the grafted PAAm improved affinity between separators and electrolyte. When the DG was too high, the pores were blocked by the PAAm, resulting in less capacity of electrolyte. Also, high amount of PAAm formed highly viscous gel matrix, the mobility of lithium ions decreased, and so did the ionic conductivity. From Fig.7, the conductivity increased with temperature, because at high temperatures, the activity of Li⁺ surged, and the conductivity of lithium ion was improved. The of PE-PAAm22.6% was the highest in conductivity, being 0.76×10⁻³ S/cm at 30℃, four times more than the pristine PE separator (0.15×10⁻³ S/cm).

Fig. 7Full-screen View

Lithium ion conductivities of pristine PE separator and PE-g-PAAm separators of different DGs, as function of temperature.

3.7. The cycling performance of lithium ion battery

Figures. 8(a) and 8(b) shows the first, tenth and twentieth cycles of charge-discharge performance of the cell assembled with PE-g-PAAm22.6% separators The first cycle specific capacity of the PE-g-PAAm cell was 121.9 mA·h/g, while it was 117.8 mA·h/g for the pristine PE cell. The charge-discharge capacities of the cells as a function of the cycle number are shown in Fig.8c, which were performed at 0.1C rate, evaluated at an operating voltage of 4.15–2.8V. After 20 cycles, the charge and discharge capacity retention ratios of the PE-g-PAAm cell were 85.3% and 93.0%, respectively, while those of the pristine PE cell were 67.9% and 73.8%, respectively. The improvements can be attributed to the PAAm gel layer, which entraps more electrolyte so that more lithium ions are yielded to anticipate electrochemical reaction, and promotes more intimate contact between electrode and separators, so as to restrain the growth of SEI film, lower the impendence on the electrode and finally suppress the lithium ion consumption [23, 24].

Fig. 8Full-screen View

Charge-discharge specific capacity of PE (a) and PE-g-PAAm22.6% (b) separators, and the cells assembled with them (c).