Gel polymer electrolytes based on nanofibrous polyacrylonitrile–acrylate for lithium batteries

G Model

MRB-7303; No. of Pages 5 Materials Research Bulletin xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Materials Research Bulletin journal homepage: www.elsevier.com/locate/matresbu

Gel polymer electrolytes based on nanofibrous polyacrylonitrile– acrylate for lithium batteries Dul-Sun Kim a, Jang Chang Woo b, Ji Ho Youk b,**, James Manuel a, Jou-Hyeon Ahn a,c,* a Department of Chemical and Biological Engineering, Research Institute for Green Energy Convergence Technology, Gyeongsang National University, 900 Gajwa-dong, Jinju 660-701, Republic of Korea b Department of Textile Engineering, Inha University, 100 Inharo, Nam-gu Incheon 402-751, Republic of Korea c Department of Materials Engineering and Convergence Technology, Gyeongsang National University, 900 Gajwa-dong, Jinju 660-701, Republic of Korea

A R T I C L E I N F O

A B S T R A C T

Article history: Available online xxx

Nanofibrous membranes for gel polymer electrolytes (GPEs) were prepared by electrospinning a mixture of polyacrylonitrile (PAN) and trimethylolpropane triacrylate (TMPTA) at weight ratios of 1/0.5 and 1/1. TMPTA is used to achieve crosslinking of fibers thereby improving mechanical strength. The average fiber diameters increased with increasing TMPTA concentration and the mechanical strength was also improved due to the enhanced crosslinking of fibers. GPEs based on electrospun membranes were prepared by soaking them in a liquid electrolyte of 1 M LiPF6 in ethylene carbonate (EC)/dimethyl carbonate (DMC) (1:1, v/v). The electrolyte uptake and ionic conductivity of GPEs based on PAN and PAN–acrylate (weight ratio; 1/1 and 1/0.5) were investigated. Ionic conductivity of GPEs based on PAN– acrylate was the highest for PAN/acrylate (1/0.5) due to the proper swelling of fibers and good affinity with liquid electrolyte. Both GPEs based on PAN and PAN–acrylate membranes show good oxidation stability, >5.0 V vs. Li/Li+. Cells with GPEs based on PAN–acrylate (1/0.5) showed remarkable cycle performance with high initial discharge capacity and low capacity fading. ß 2014 Elsevier Ltd. All rights reserved.

Keywords: A. Nanostructures C. Electron microscopy C. Impedance spectroscopy D. Electrochemical properties D. Ionic conductivity

1. Introduction The majority of lithium secondary batteries currently use liquid electrolytes, which can easily leak out of the cells and can cause safety problems. Lithium polymer batteries employing gel or solid polymer electrolytes have attracted much attention and various polymer electrolytes have been studied to develop technology of lithium polymer batteries and enhance the electrochemical performance of cells since they possess a great variety of advantages such as excellent safety characteristics, flexibility of shape and high ionic conductivity [1]. Gel polymer electrolytes (GPEs) are usually made from a polymer host, a salt and a solvent or a mixture of solvents. GPEs have good ionic conductivity with relatively good mechanical properties at a wide temperature range due to the inclusion of a higher amount of organic electrolyte in the polymer hosts.

* Corresponding author at: Department of Chemical and Biological Engineering, Research Institute for Green Energy Convergence Technology, Gyeongsang National University, 900 Gajwa-dong, Jinju 660-701, Republic of Korea. Tel.: +82 55 772 1784; fax: +82 55 772 1789. ** Corresponding author. Tel.: +82 32 860 7498; fax: +82 32 873 0181. E-mail addresses: [email protected] (J.H. Youk), [email protected] (J.-H. Ahn).

Many polymers can be used in GPEs, such as poly(ethylene oxide) (PEO) [2–6], poly(ethylene glycol) (PEG) [7], poly(vinylidene fluoride) (PVdF) [8–10], poly(vinylidene fluoride-co-hexafluoropropylene) (P(VdF-HFP)) [11–14], and polyacrylonitrile (PAN) [15– 17]. Among these polymer hosts, PAN-based GPEs offer many good characteristics like high ionic conductivity, thermal stability, good morphology for electrolyte uptake and compatibility with lithium electrodes [18]. Also, PAN can minimize the formation of dendrite growth during the charging/discharging process of lithium-ion polymer batteries [16]. However, PAN-based GPEs suffer from poor mechanical strength that is difficult to meet the requirement of practical application of lithium polymer batteries [19,20]. Therefore many efforts have been made to improve PAN-based GPEs. Rajendran et al. [19] reported GPEs based on PAN–PVC prepared by casting technology. They demonstrated that the addition of PVC could improve the mechanical strength of polymer membranes and achieve strong and free-standing thin films. However, the GPEs based on PAN–PVC had a low ionic conductivity. Prasanth et al. [21] reported electrospinning is an efficient method to prepare polymer nanofibrous membrane with fully interconnected pore structure and the membranes show high porosity and electrolyte uptake, and comparable leakage property compared to conventional separator. PAN–PVC fibrous membranes were prepared by

http://dx.doi.org/10.1016/j.materresbull.2014.01.047 0025-5408/ß 2014 Elsevier Ltd. All rights reserved.

Please cite this article in press as: D.-S. Kim, et al., Mater. Res. Bull. (2014), http://dx.doi.org/10.1016/j.materresbull.2014.01.047

G Model

MRB-7303; No. of Pages 5 D.-S. Kim et al. / Materials Research Bulletin xxx (2014) xxx–xxx

2

Zhong et al. [22] with electrospinning and such electrospun PAN– PVC GPEs showed both good mechanical stability and electrochemical properties. Trimethylolpropane triacrylate (TMPTA) was reported for use in the preparation of GPEs with a crosslinking structure, and the cell with TMPTA-based GPEs showed stable cycle performance [23]. In this work, naonfibrous membranes for GPEs were prepared by blending PAN with TMPTA to form crosslinking of fibers thereby improving their mechanical strength. The physical and electrochemical characteristics of the GPEs were studied in detail. 2. Experimental PAN is commercially purchased from Polyscience Inc. with a number-average molecular weight (Mn) of 150,000 g/mol. It was vacuum dried at 60 8C for 6 h before use. Trimethylolpropane triacrylate (TMPTA, Aldrich) and azobisisobutyronitrile (AIBN, Duksan) were used as the crosslinking agent and initiator, respectively. The solvent N,N-dimethyl formamide (DMF, Aldrich) was used as received. Nanofibrous membranes were prepared by a typical electrospinning method at room temperature [11,12]. PAN and TMPTA with different weight ratios (1.0/0.5 and 1.0/1.0 g/g) were homogeneously dissolved in 10 mL of DMF by mechanical stirring at 60 8C. 0.02 g of AIBN was added to the solution with stirring. The resulting solutions were electrospun at 25 8C and the essential spinning parameters used are: applied voltage 15 kV, distance between the tip of the needle and collecting Plate 20 cm, needle size 1.27 mm and solution feed rate 2.0 mL/h. The electorspun membranes thus obtained were then annealed for radical reaction at 100 8C for 3 h. PAN membranes without TMPTA were also electrospun using a 13 wt% PAN solution in DMF for comparison. The fiber morphology was observed with scanning electron microscope (SEM, JEOL JSM 5600) and the average fiber diameter (AFD) was estimated from the micrographs taken at high magnification. The tensile strength of PAN and PAN–acrylate membranes were evaluated with universal testing machine (UTM, Instron 3343). The porosity of the membranes was measured by mercury porosimetry (Micromeritics, ASAP 2010). GPEs were prepared by soaking the electrospun nanofibrous membranes in the electrolyte solution: 1 M LiPF6 in ethylene carbonate (EC)/dimethyl carbonate (DMC) (1:1, v/v) (Soulbrain Co., Ltd.) mixture. The electrolyte uptake was calculated according to the following equation:

uptake ð%Þ ¼

Wt  W0  100; W0

where W0 is the mass of the dry membrane and Wt is the mass of the membrane after soaking with the electrolyte. Ionic conductivity of GPEs was measured by the AC impedance of the cell with 1M6 frequency analyzer over the temperature range from 0 to 60 8C. The cell was fabricated by sandwiching GPEs between two stainless steel electrodes (SS/GPE/SS) and the impedance measurements were performed at an amplitude of 20 mV over the frequency

range from 100 mHz to 2 MHz. The cell was kept at each measuring temperature for 1 h to ensure the thermal equilibration of the sample at temperature before measurement. The ionic conductivity could be calculated based on the following equation:

s¼

t  A; Rb

where s is the ionic conductivity (S cm1), Rb is the bulk resistance (V), t and A are the thickness (cm) and area (cm2) of the GPEs, respectively. Electrochemical stability was determined by linear sweep voltammetry (LSV) using Li/GPE/SS cells at a scanning rate of 1 mV s1 over the potential range of 2.5–6 V vs. Li/Li+. Swagelok cells were fabricated by placing the GPEs between lithium metal anode and carbon coated lithium iron phosphate (LiFePO4) cathode. LiFePO4 was synthesized as previously studied [24] and the cathode was prepared by mixing LiFePO4 powder, conductive carbon and PVdF binder in the weight ratio of 80:10:10 with N-methyl-2-pyrrolidone (NMP). The slurry was cast onto aluminum foil and dried at 80 8C for 24 h in vacuum for further use. The electrochemical tests of the Li/GPE/LiFePO4 cells were performed in an automatic galvanostatic charge–discharge unit, WBCS3000 battery cycler (WonA Tech. Co.), between 2 and 4.2 V at room temperature with a current density of 0.1 C. 3. Results and discussion The preparation of nanofibrous membranes based on PAN– acrylate is aimed to achieve freestanding GPEs with good mechanical strength by the addition of TMPTA as a crosslinking agent. As expected, the mechanical properties of PAN–acrylate electrospun membranes are improved compared to PAN membranes as given in Table 1. Both tensile strength and modulus are higher for PAN–acrylate membranes. The morphology of electrospun nanofibrous PAN and PAN– acrylate membranes is shown in Fig. 1(a–c). All membranes showed a three-dimensional network structure with fully interconnected pores made up of fibers. The average fiber diameter is around 435 nm for PAN membranes, and 522 and 620 nm for PAN– acrylate (1/0.5) and PAN–acrylate (1/1), respectively. PAN–acrylate (1/0.5) membranes were observed to have denser morphology with smaller pores than PAN membranes, which is supposed to result from the addition of TMPTA that formed a crosslinking structure. It was found that the fiber diameter was not uniform with increased concentration of TMPTA (Fig. 1(c)), which suggests that an excess of TMPTA in the polymer blend may increase the viscosity of the polymer blend solution with too much crosslinking agent that distort the uniformity of fibers. To observe the swelling behavior of electrospun membranes, they were soaked in liquid electrolyte of 1 M LiPF6 in EC/DMC (1/1, v/v) for 15 h. Membranes were then washed with ethanol and dried at 80 8C in vacuum for 12 h before taking SEM observation. The SEM images of nanofibrous PAN and PAN–acrylate membranes after being soaked in liquid electrolyte for 15 h are given in Fig. 1(d–f). It is shown that the swelling ability was much improved

Table 1 Properties of electrospun membranes and GPEs based on nanofibrous PAN and PAN–acrylate membranes activated with 1 M LiPF6 in EC/DMC. Properties

PAN

PAN–acrylate (1/0.5, w/w)

PAN–acrylate (1/1, w/w)

Average fiber diameter (nm) Tensile strength (MPa) Modulus (MPa) Porosity (%) Thickness (um) Electrolyte uptake (%) Ionic conductivity at 25 8C (mS cm1)

435 13.0 454.0 78.8 45 420 3.77

522 14.2 556.1 72.8 40 340 5.22

620 15.4 661.0 64.1 33 280 2.59

Please cite this article in press as: D.-S. Kim, et al., Mater. Res. Bull. (2014), http://dx.doi.org/10.1016/j.materresbull.2014.01.047

G Model

MRB-7303; No. of Pages 5 D.-S. Kim et al. / Materials Research Bulletin xxx (2014) xxx–xxx

3

Fig. 1. SEM images of PAN (a), PAN–acrylate (1/0.5) (b) and PAN–acrylate (1/1) (c) and GPEs based on PAN (d), PAN–acrylate (1/0.5) (e), PAN–acrylate (1/1) (f) soaked in 1 M LiPF6 in EC/DMC for 15 h.

by adding TMPTA, which is expected to make GPEs in cells stabilize faster and absorb much more electrolyte. However, PAN–acrylate (1/1) membrane was swollen so much that the pores were partially blocked, which may hinder Li+ movement. PAN–acrylate (1/0.5) membranes were shown to maintain the interconnected network while swelling, indicating a good affinity with electrolyte solution due to the polar functional groups in PAN–acrylate [25,26]. Electrolyte uptake of electorspun membranes in 1 M LiPF6 in EC/DMC were tested and are shown in Fig. 2. The difference in electrolyte uptake is attributed to the difference in morphology, packing of fibers, porosity and pore structure. The fully interconnected pore structure of electrospun membranes allows fast penetration of the liquid electrolyte into the membranes, and hence the uptake process is stabilized within only 10 min for PAN and PAN–acrylate membranes. PAN–acrylate membranes showed lower electrolyte uptake than PAN membranes, which may be due

500

Electrolyte uptake (%)

400

300

200

PAN PAN-acrylate (1/0.5) PAN-acrylate (1/1)

100

0

0

10

20

30

40

50

60

70

Wetting time (min) Fig. 2. Electrolyte uptake of nanofibrous PAN and PAN–acrylate membranes (liquid electrolyte: 1 M LiPF6 in EC/DMC (1/1)).

to the densely packed fibers with smaller pores in PAN–acrylate membranes. Fig. 3(a) shows the AC impedance spectra of GPEs based on nanofibrous PAN and PAN–acrylate membranes at 25 8C. The straight lines inclined toward the real axis represent the electrode/ electrolyte double layer capacitance behavior. The bulk resistance (Rb) was obtained from the intercept on the real axis. From the impedance data, the ionic conductivities of GPEs at 25 8C are calculated and listed in Table 1. PAN–acrylate (1/0.5) membranes achieved the highest ionic conductivity of 5.22 mS cm1. This can be ascribed to the good affinity of membranes and liquid electrolyte. PAN–acrylate (1/1) membranes exhibited the lowest ionic conductivity, which may result from the low porosity of the membrane with significant swelling behavior of the fibers themself. The temperature-dependent ionic conductivity of GPEs in the range of 0–60 8C is shown in Fig. 3(b). The plot is approximately linear, suggesting an Arrhenius-like behavior [27]. It can be observed that the ionic conductivity increases with increasing temperature since higher temperature promotes ions to move faster. The anodic stability window of GPEs based on PAN and PAN– acrylate membranes is shown in Fig. 4. All GPEs exhibit stabilities greater than 5.0 V vs. Li/Li+ and GPEs based on PAN–acrylate (1/0.5) membranes are stable up to 5.2 V vs. Li/Li+. The GPEs based on PAN and PAN–acrylate membranes have been evaluated for chargedischarge performance in Li/LiFePO4 cells at room temperature. The initial charge–discharge curves and cycle performance of Li/LiFePO4 cell with GPEs studied here at current density 0.1 C are presented in Fig. 5(a). Cells with GPEs based on PAN, PAN–acrylate (1/0.5) and PAN–acrylate (1/1) membranes delivered initial discharge capacity of 140 mAh g1, 145 mAh g1 and 138 mAh g1, respectively. It was found that the discharge capacity of the cell with GPEs based on PAN/ acrylate (1/0.5) achieved the highest initial capacity corresponding to 82.3% of the theoretical capacity, which is due to the highest ionic conductivity. The cycle performance of the cells up to 50 cycles is shown in Fig. 5(b). Cells with GPEs based on PAN/acrylate (1/0.5) retained 96.5% of initial discharge capacities. The good cycle property of the cell can be attributed to the improved ionic conductivity, mechanical property and good compatibility between

Please cite this article in press as: D.-S. Kim, et al., Mater. Res. Bull. (2014), http://dx.doi.org/10.1016/j.materresbull.2014.01.047

G Model

MRB-7303; No. of Pages 5 D.-S. Kim et al. / Materials Research Bulletin xxx (2014) xxx–xxx

4

8

4.5

PAN PAN/acrylate(1/0.5) PAN/acrylate(1/1)

7 6

Voltage (V)

3.5

- Z ( Ω)

5 4

(a)

3

1

2

3

4

5

6

7

Discharge Capacity (mAh/g)

PAN PAN/acrylate(1/0.5) PAN/acrylate(1/1)

-2.0 -1

40

60

80

100

120

140

160

160

-1.8

-2.2 -2.4 -2.6

(b)

-2.8

140 120

3.0

3.2

3.4 -1

3.6

(b)

100 80 60

PAN PAN-acryl(1/0.5) PAN-acryl(1/1)

40 20 0

-3.0 2.8

0

5

10

15

20

25

30

35

40

45

50

Cycle Number

3.8

-1

1000T (K ) Fig. 3. AC impedance spectra at 25 8C (a) and temperature-dependent ionic conductivity (b) of GPEs based on nanofibrous PAN and PAN–acrylate membranes.

Fig. 5. Initial charge–discharge properties (a) and cycle performance (b) of Li/GPE/ LiFePO4 cells with GPEs based on nanofibrous PAN and PAN–acrylate membranes (25 8C, 0.1 C-rate, 2–4.2 V).

the electrolyte and electrodes. The results illustrate that GPEs based on PAN–acrylate membranes are suitable for applications in lithium secondary batteries.

0.30

PAN PAN-acrylate (1/0.5) PAN-acrylate (1/1)

4. Conclusions

0.20

2

Current (mA/cm )

20

8

Z (Ω)

0.15 0.10 0.05 0.00 2.5

0

Specific capacity (mAh/g) 0

-1.6

0.25

PAN PAN-acryl(1/0.5) PAN-acryl(1/1)

2.5

1.5

1

Log σ (Scm )

3.0

2.0

2

0

(a)

4.0

3.0

3.5

4.0

4.5

5.0

5.5

6.0

+

Voltage (V vs. Li/Li ) Fig. 4. Anodic stability windows of GPEs based on nanofibrous PAN and PAN– acrylate membranes.

GPEs based on nanofibrous PAN and PAN–acrylate membranes were prepared by electrospinning of a polymer blend solution of PAN and TMPTA in DMF. TMPTA was used as the crosslinking agent and it helps to improve the physical property and swelling behavior of membranes in liquid electrolyte. It is found that mechanical strength was improved by the addition of TMPTA and GPEs based on PAN–acrylate (1/0.5) showed the highest ionic conductivity up to 5.22 mS cm1 due to the proper swelling behavior of fibers and good affinity with liquid electrolyte. However, excess TMPTA results in distort morphology and low ionic conductivity. Cells fabricated with GPEs show good oxidation stability >5.0 V. Swagelok cells using GPEs based on PAN and PAN– acrylate membranes with 1 M LiPF6 in EC/DMC were evaluated with LiFePO4 cathode. Remarkable cycle performance was obtained with high initial discharge capacity and low capacity fading with continuous cycling.

Please cite this article in press as: D.-S. Kim, et al., Mater. Res. Bull. (2014), http://dx.doi.org/10.1016/j.materresbull.2014.01.047

G Model

MRB-7303; No. of Pages 5 D.-S. Kim et al. / Materials Research Bulletin xxx (2014) xxx–xxx

Acknowledgements This work was supported by the Industrial Strategic Technology Development Program (10040033) funded by the Ministry of Trade, Industry & Energy (MOTIE) and also supported by the National Research Foundation of Korea (NRF) grant funded by the Ministry of Science, ICT & Future Planning (NRF2013R1A2A2A04016075). References [1] D.H. Lim, J. Manuel, J.H. Ahn, J.K. Kim, P. Jacobsson, A. Matic, J.K. Ha, K.K. Cho, K.W. Kim, Solid State Ion. 225 (2012) 631–635. [2] H. Miyashiro, S. Seki, Y. Kobayashi, Y. Ohno, Y. Mita, A. Usami, Electrochem. Commun. 7 (2005) 1083–1086. [3] Y. Kobayashi, S. Seki, A. Yamanaka, H. Miyashiro, Y. Mita, T. Iwahori, J. Power Sources 146 (2005) 719–722. [4] P. Reale, S. Panero, B. Scrosati, J. Garche, M. Wohlfahrt-Mehrens, M. Wachtler, J. Electrochem. Soc. 151 (2004) A2138–A2142. [5] Y.K. Kang, J. Lee, D.H. Suh, C. Lee, J. Power Sources 146 (2005) 391–396. [6] J.W. Choi, G. Cheruvally, Y.H. Kim, J.K. Kim, J. Manuel, P. Raghavan, J.H. Ahn, K.W. Kim, H.J. Ahn, D.S. Choi, C.E. Song, Solid State Ion. 178 (2007) 1235–1241. [7] N.T.K. Sundaram, A. Subramania, Electrochim. Acta 52 (2007) 4987–4993. [8] C.Y. Chiang, M.J. Reddy, P.P. Chu, Solid State Ion. 175 (2004) 631–635. [9] Y.J. Wang, D. Kim, Electrochim. Acta 52 (2007) 3181–3189. [10] V. Gentili, S. Panero, P. Reale, B. Scrosati, J. Power Sources 170 (2007) 185–190.

5

[11] P. Raghavan, J.W. Choi, J.H. Ahn, G. Cheruvally, G.S. Chauhan, H.J. Ahn, A. Nah, J. Power Sources 184 (2008) 437–443. [12] P. Raghavan, X. Zhao, J.K. Kim, J. Manuel, G.S. Chauhan, J.H. Ahn, C. Nah, Electrochim. Acta 54 (2008) 228–234. [13] P. Raghavan, X. Zhao, J. Manuel, G.S. Chauhan, J.H. Ahn, H.S. Ryu, H.J. Ahn, K.W. Kim, C. Nah, Electrochim. Acta 55 (2010) 1347–1354. [14] J.K. Kim, G. Cheruvally, X. Li, J.H. Ahn, K.W. Kim, H.J. Ahn, J. Power Sources 178 (2008) 815–820. [15] H.S. Min, J.M. Ko, D.W. Kim, J. Power Sources 119 (2003) 469–472. [16] A.I. Gopalan, P. Santhosh, K.M. Manesh, J.H. Nho, S.H. Kim, C.G. Hwang, K.P. Lee, J. Membr. Sci. 325 (2008) 683–690. [17] P. Raghavan, J. Manuel, X. Zhao, D.S. Kim, J.H. Ahn, C. Nah, J. Power Sources 196 (2011) 6742–6749. [18] H. Tsutsumi, A. Matsuo, K. Takase, S. Doi, A. Hisanaga, K. Onimura, T. Oishi, J. Power Sources 90 (2000) 33–38. [19] S. Rajendran, R. Babu, P. Sivakumar, J. Appl. Polym. Sci. 113 (2009) 1651–1656. [20] S. Rajendran, R. Babu, P. Sivakumar, Ionics 14 (2008) 149–155. [21] P. Raghavan, X. Zhao, J. Manuel, C. Shin, M.Y. Heo, J.H. Ahna, H.S. Ryu, H.J. Ahn, J.P. Noh, G.B. Cho, Mater. Res. Bull. 45 (2010) 362–366. [22] Z. Zhong, Q. Cao, B. Jing, S. Li, X. Wang, Ionics 18 (2012) 853–859. [23] S.I. Kim, H.S. Kim, S.H. Na, S.I. Moon, Y.J. Kim, N.J. Jo, Electrochim. Acta 50 (2004) 317–321. [24] J.K. Kim, J.W. Choi, G. Cheruvally, J.U. Kim, J.H. Ahn, G.B. Cho, K.W. Kim, H.J. Ahn, Mater. Lett. 61 (2007) 3822–3825. [25] J. Saunier, F. Alloin, J.Y. Sanchez, G. Caillon, J. Power Sources 119–121 (2003) 454–459. [26] J.R. Kim, S.W. Choi, S.M. Jo, W.S. Lee, B.C. Kim, J. Electrochem. Soc. 152 (2005) A295–A300. [27] Y.X. Jiang, Z.F. Chen, Q.C. Zhuang, J.M. Xu, Q.F. Dong, L. Huang, S.G. Sun, J. Power Sources 160 (2006) 1320–1328.

Please cite this article in press as: D.-S. Kim, et al., Mater. Res. Bull. (2014), http://dx.doi.org/10.1016/j.materresbull.2014.01.047