Russian Journal of Electrochemistry, Vol. 37, No. 3, 2001, pp. 280–286. Translated from Elektrokhimiya, Vol. 37, No. 3, 2001, pp. 321–327. Original Russian Text Copyright © 2001 by Yarmolenko, Belov, Efimov.

Effect of Crown Ethers on the Conduction of Plasticized Polyacrylonitrile-Based Electrolytes O. V. Yarmolenko, D. G. Belov, and O. N. Efimov Institute of Problems of Chemical Physics, Russian Academy of Sciences, p/o Chernogolovka, Noginsk raion, Moscow oblast, 142432 Russia Received March 14, 2000

Abstract—A series of new plasticized electrolytes based on a lithium salt, polyacrylonitrile, propylene carbonate, and such crown ethers as 15-crown-5 and benzo-15-crown-5 as additives is synthesized and studied. According to impedance spectroscopy, the electrolytes’ conductivity is 6 × 10–3 S cm–1 at room temperature. The electrolytes’ compatibility with a new thin-film material (polyacetylene-covered porous polypropylene, which is used for protecting lithium anodes) is investigated.

INTRODUCTION Solid-polymer electrolytes (SPE) in lithium batteries have some advantages over organic electrolytes. These enhance fire- and explosion-safety and simplify the technology used for assembling batteries of any shape, for example, planar. Modifying a polyethyleneoxide-based solid electrolyte with crown ethers increases its conductivity up to 10–3 S cm–1 [1]. However, the problem of high polarization at the SPE/lithium electrode interface, which is due to poor mechanical contact between these, has not been solved yet. To increase the electrolyte conductivity and decrease the resistance of the SPE/electrode interface, SPE may be plasticized by an aprotic organic solvent with a high bp and a high dielectric constant [2, 3]. We prepared and studied a series of plasticized polymer electrolytes which include propylene carbonate (PC) as organic solvent and a crown ether as additive, for improving the contact and decreasing the polarization. However, in doing so, it was desirable to reduce decomposition of the solvent interacting with lithium, which is especially strongly pronounced in the course of the lithium deposition during charging [4]. This gives rise to a poorly conducting salt layer, and a portion of lithium forms isolated metallic islets. In turn, this reduces the active lithium mass that takes part in charge–discharge cycles and eventually defines the negative-electrode capacity. The lithium electrode surface can efficiently be protected from destruction by covering it with a polyacetylene-based polymer coating that has both electron conduction and ionic conduction by lithium cations [5]. When passing through such a coating, the lithium ion undergoes desolvation, thus eliminating immediate contact between lithium and the solvent (Fig. 1). Although the polyacetylene coating has electron conduction, self-discharge does not occur. We explain these experimental facts by assuming that lithium

exists in polyacetylene (PA) in the form of cations and there are no active centers on PA, which would catalyze the electrolyte decomposition. Moreover, the coating inhibits the dendrite growth, thus reducing the probability of internal shorts. During cathodic polarization, lithium cations are intercalated into PA, which acquires a positive charge and becomes highly conducting [6]: PA + δe + δLi+

PAδ–δLi+.

Lithium cations easily diffuse through the thin (5– 10 µm) polyacetylene coating and then discharge at the surface of metallic lithium: Li+ + e

Li.

These electrode reactions are reversible. During a battery discharge, lithium undergoes ionization; in the form of cation, diffuses through the polyacetylene coating to the boundary with electrolyte; and then passes into electrolyte. Unfortunately, free polyacetylene films are inconvenient to use, as they relatively easily oxidize in air and rapidly become brittle. Therefore, it was necessary to improve physicochemical properties of polyacetylene and increase its endurance limit. To this end, it was proposed that the separator material on the basis of porous propylene (PP) be filled in by polymerizing acetylene in the pores and on the surface [7]. EXPERIMENTAL Electrolyte Components Polyacrylonitrile (PAN), which is a white powder of difficultly crystallizing polymer with a melting point of 86–96°C and a molecular weight of 120000, was used as received. The LiClO4 salt was dried in a vacuum for two hours at 120°ë and for five hours at 160°ë. Propylene carbonate (bp of 240°C) was stored over molecular sieves 4 Å,

1023-1935/01/3703-0280$25.00 © 2001 åÄIä “Nauka /Interperiodica”

EFFECT OF CROWN ETHERS

Li+

281

PE

PP

PA

Li

Fig. 1. Schematics of half-cell lithium/polyacetylene on polypropylene electrolyte.

then filtered and distilled in a vacuum of 2.5 mmHg at a temperature of 78°ë, in the presence of lithium shavings. Benzo-15-crown-5 (pinkish crystalline powder with mp of about 78–80°C) was recrystallized from isopropyl alcohol, while 15-crown-5 (colorless liquid) was distilled in a vacuum of 2 mmHg at bp of 120– 122°C. Synthesis of Polymer Electrolytes To synthesize a polymer electrolyte (PE), we placed a weight of polyacrylonitrile into a flask with a vacuum cock and dried it at 50°ë for an hour. Due to a premature formation of cross-linkages in polyacrylonitrile by the C≡N bond, a temperature elevation is prohibitive at this stage. Then, LiClO4, propylene carbonate, and (in selected experiments) a crown ether were added into the flask under argon. The mixture in the sealed flask filled with argon was heated to a temperature of 150°ë, at which cross-linkages form in polyacrylonitrile in accordance with the scheme [8] [CH2 CH2]n CN

150°C

CH2 CH2 CH CH .

CH C

C N

C N

RUSSIAN JOURNAL OF ELECTROCHEMISTRY

Vol. 37

The heating continued until a homogeneous mass formed. Afterwards, the flask was kept at room temperature for 24 hours. The plasticized electrolyte thus prepared (a yellowish rubbery mass) was pressed between two electrodes made of a lithium band 1 mm thick. The PE thickness was 1.7 to 2.0 mm. The PE compositions were as follows: (a) PAN 10, LiClO4 3.5, and PC 86.5; (b) PAN 8.5, LiClO4 3.5, PC 78, and benzo-15-crown5 10; and (c) PAN 8.5, LiClO4 3.0, PC 81, and 15crown-5 7.5 (ingredient amounts are given in wt %). Polyacetylene Synthesis on Polypropylene The separator material was a microporous band of non-woven polypropylene (PP) 20–25 µm thick, with a porosity of 53% and an effective pore diameter of 0.03 µm. The material was intended for use in batteries with aqueous and organic electrolytes. The PP strength exceeded 1000 kg cm–2. When polymerizing acetylene, we employed a Shirakawa catalyst [9] containing Ti(OBu)4 and AlEt3 in a 1 : 4 molar ratio (AlEt3 concentration 0.115 g ml–1). The components were placed in an ampoule filled with toluene and flushed with argon and then stirred with a magnetic stirrer for 30–60 min at room temperature. The prepared catalyst was used to treat a PP strip and then vacuumed, in order to remove the solvent and conNo. 3

2001

282

YARMOLENKO et al.

1 µm (a)

1 µm (b) Fig. 2. Electron micrographs of surfaces of (a) initial PP film and (b) composite PA–PP, taken with JSM-53LV microscope (JEOL); magnification 10000.

centrate the catalyst. The acetylene polymerization was performed at room temperature for 10–60 min. Then the strip was washed with toluene to remove the catalyst residues and dried in a vacuum. At the final stage, the PP strip with a one-side PA coating was rolled onto a lithium electrode with the PA-coated side facing the electrode. In the process, the low-density porous material compacted a little and sufficiently strongly adhered to the metal surface. The initial PP separator has oval pores a few micrometers long. The pore width does not exceed 1 µm (Fig. 2a). The walls of the pores are connected by cords. The structure of the strip cross-sections was studied by cleaving it in liquid nitrogen. The electron

micrographs show a uniform distribution of pores over the volume in directions parallel and perpendicular to the extrusion axis (Fig. 2b). The structure of composite materials was more monolithic. Pores partially remained in the film bulk. It is impossible to discern pores in PA. However, they are likely to exist in a considerable amount, because the BET measurements show that the surface area increases when coated with PA. As opposed to porous polypropylene, polyacetylene probably has a system of micropores, which is likely to favor the intercalation and transport of lithium ions deprived of their solvation shells. Mechanical and electrophysical properties of obtained composite materials are superior to those of porous polypropylene and

RUSSIAN JOURNAL OF ELECTROCHEMISTRY

Vol. 37

No. 3

2001

EFFECT OF CROWN ETHERS

283

free polyacetylene. The composite materials are readily pressed onto metallic surfaces (stainless steel, lithium).

Cdl

Instruments The electrochemical impedance was measured with a Solartron 1255 frequency analyzer at frequencies of 1 to 105 Hz. The amplitude of the measuring signal was 5–10 mV. The results of the measurements were processed in the model of adsorption relaxation of the double layer [10], using the equivalent circuit depicted in Fig. 3.

Re

RF

RA

RESULTS AND DISCUSSION Introducing a crown ether into a polyacrylonitrilebased porous electrolyte improves the electrolyte’s conduction owing to the complexing between the crown ether and the lithium ion. The formation of such complexes may weaken the ionic bond in LiClO4, thus increasing mobility of lithium ions. We prepared and studied some new porous electrolytes based on polyacrylonitrile, LiClO4, propylene carbonate, and such crown ethers as 15-crown-5 (15C5) and benzo-15crown-5 (B15C5). For studying the electrolytes’ conduction, we employed the impedance spectroscopy method. We used four types of symmetrical electrochemical cells. First, cells with blocking stainless-steel electrodes (ss). Second, cells with reversible lithium electrodes (Li). Third, cells with thin conducting PA– PP films pressed onto stainless-steel current leads (ss/PA–PP). And finally, cells with reversible lithium electrodes covered by PA–PP films (Li/PA–PP). We studied not only the way the PE bulk conduction depends on the PE composition. Simultaneously, we investigated the conduction of the faradaic reaction Li+ + e Li at the electrolyte/electrode interface (Table 1). As seen in Table 1, the conductivity of the system with irreversible electrodes increases threefold after introducing 15-crown-5. No increase in the conductiv-

CA

ZW

Fig. 3. Equivalent circuit, where Re is bulk resistance of PE, RF is charge transfer resistance, Cdl is double-layer capacitance, RA is adsorption resistance, CA is adsorption capacitance, and ZW is diffusion impedance.

ity occurs in the system with lithium electrodes. We believe that the crown ether has no impact on the latter system because its concentration in the electrolyte bulk decreases. In a plasticized electrolyte, the crown ether freely moves towards the lithium surface through PCfilled channels, where it is energetically favorable for it to stay because of a large excess of lithium ions. In the cells with blocking stainless-steel electrodes, the resistance of the faradaic process reaches as much as 1010 to 1012 ohm, and we do not discuss this fact below. In systems with electrodes covered by thin PA–PP films, the bulk conductivity σsp of the PA–PP/PE/PA– PP system is lower than that of PE by an order of magnitude (Table 2). The faradaic conductivity of the PA– PP/Li interface is lower than that of the PE/Li interface also by a factor of ten. As we have already pointed out, the film PA–PP material is conducting by lithium ions only via a system of through pores filled with PA, rather than over the entire surface. This is precisely what leads to a decrease in the conductivity by lithium ions. The

Table 1. Electrolyte conductivity and “faradaic conductivity” at the PE/electrode interface at 20°C Electrochemical system (1) ss/PE/ss (2) Li/PE/Li

PAN, LiClO4, PC σsp , S

σF , S

cm–1

1.9 × 10–3 1.8 × 10–3

PAN, LiClO4, PC, 15C5 cm–2

σsp , S cm–1

σF , S cm–2

<10–10 1.5 × 10–3

6.3 × 10–3 1.7 × 10–3

<10–10 1.4 × 10–3

Table 2. Conductivity of the PA–PP/PE/PA–PP system and “faradaic conductivity” at the PA–PP/electrode interface at 20°C Electrochemical system (1) ss/PA–PP/PE/PA–PP/ss (2) Li/PA–PP/PE/PA–PP/Li

PAN, LiClO4, PC

PAN, LiClO4, PC, 15C5

σsp , S cm–1

σF , S cm–2

σsp , S cm–1

σF , S cm–2

2.4 × 10–4 1.1 × 10–4

<10–10 2.6 × 10–4

5.0 × 10–4 6.1 × 10–4

<10–10 1.0 × 10–4

RUSSIAN JOURNAL OF ELECTROCHEMISTRY

Vol. 37

No. 3

2001

284

YARMOLENKO et al.

Table 3. Conductivity values at different temperatures σsp , S cm–1

Electrochemical system (1) Li/PAN, LiClO4, PC/Li (2) Li/PAN, LiClO4, PC, B15C5/Li (3) Li/PAN, LiClO4, PC, 15C5/Li

–10°C

18°C

50°C

0.8 × 10–3 0.6 × 10–3 1.3 × 10–3

1.8 × 10–3 1.3 × 10–3 1.7 × 10–3

2.8 × 10–3 1.9 × 10–3 3.6 × 10–3

Table 4. Faradaic conductivity of the electrode/electrolyte interface at different temperatures σF , S cm–1

Electrochemical system (1) Li/PAN, LiClO4, PC/Li (2) Li/PAN, LiClO4, PC, B15C5 (3) Li/PAN, LiClO4, PC, 15C5/Li

–10°C

18°C

50°C

3.1 × 10–4 2.5 × 10–4 1.8 × 10–4

1.5 × 10–3 1.2 × 10–3 1.4 × 10–3

3.8 × 10–2 3.5 × 10–2 1.6 × 10–2

Ea , eV 0.90 0.83 0.62

Table 5. Conductivity of the Li/PE/Li cells as a function of storage time t t=0

Electrolyte composition (1) PAN, LiClO4, PC (2) PAN, LiClO4, PC, B15C5 (3) PAN, LiClO4, PC, 15C5

t = 16 months

σsp , S cm–1

σF , S cm–2

σsp , S cm–1

σF , S cm–2

1.8 × 10–3 1.3 × 10–3 1.7 × 10–3

1.5 × 10–3 1.2 × 10–3 1.4 × 10–3

2.1 × 10–3 1.5 × 10–3 3.6 × 10–3

0.5 × 10–3 1.0 × 10–3 3.3 × 10–3

conductivity-enhancing effect of a crown ether additive remains intact. Our investigation of the Li/PE/Li system was more thorough. We found out how the composition and tem-

perature (–10, 18, 50°C) affect the bulk and faradaic conductivities of PE (Tables 3, 4). The complex-plane plots for the Li/PE/Li cell impedance are presented in Figs. 4 and 5 for each cell type.

–Im, ohm

40

1 kHz

20 10 kHz

10 kHz 1 Hz

100 kHz

–Im, ohm

1 2 3

0

1000

60

100 kHz

120 Re, ohm

10 Hz 1 Hz

500

100 Hz 10 Hz 1 Hz

0

1000

2000

3000 Re, ohm

Fig. 4. Impedance spectra for the Li/PAN, LiClO4, PC/Li cell at (1) –10, (2) 20, and (3) 50°C. RUSSIAN JOURNAL OF ELECTROCHEMISTRY

Vol. 37

No. 3

2001

EFFECT OF CROWN ETHERS –Im, ohm 1600

10 Hz

1

285

–Im, ohm 1 800 2

2

100 Hz

10 Hz

100 Hz

10 Hz

100

3 1200

1 Hz 100 kHz

0

100

200

300

400

500

600 Re, ohm

800 10 Hz

Fig. 6. Impedance spectra for the Li/PAN, LiClO4, PC, 15C5/Li cell: (1) initial and (2) 16 months later.

100 Hz

400 100 Hz 100 Hz 1 Hz

0

400

800

1200

1600 Re, ohm

Fig. 5. Impedance spectra for the Li/PAN, LiClO4, PC, 15C5/Li cell at (1) –10, (2) 20, and (3) 50°C.

which evidences a positive effect of this crown ether. By way of example, Fig. 6 shows complex-plane plots for the impedance of the cell based on PE containing 15-crown-5. After storage of these cells, the bulk conductivity of PE containing crown ether reduces by half (Table 6). Due to the electrode irreversibility, we present no data on the faradaic conductivity of the PA/PP/ss interface. The cell with PE free of crown ether is stable. CONCLUSION

The PE conductivity depends only weakly on the temperature (Table 3); hence, lithium batteries on the basis of PE may operate in the temperature range –10 to 50°C. With the temperature changed by 30°ë, the rate of the faradaic reaction at the PE/Li interface increases by an order of magnitude (Table 4). The temperature dependence of the logarithm of the electron transfer rate is linear in the Arrhenius coordinates. Table 4 also contains calculated activation energies Ea. In the presence of 15-crown-5, the rate of reaction Li+ + e Li is maximum. In addition to the temperature dependence for the Li/PE/Li cells, we studied the time dependence of the resistance of these cells and of the ss/PA–PP/PE/PA-PP/ss systems (Table 5). As seen in Table 5, after a 16-month storage of the cells, the bulk conductivity of electrolytes free of crown ethers undergoes no change, whereas the PE/Li interface conductivity drops by a factor of three. The electrolyte containing benzo-15-crown-5 is stable. Introducing 15-crown-5 into PE leads to an increase in both the bulk conductivity and the faradaic conductivity, Table 6. Conductivity of the ss/PA–PP/PE/PA–PP/ss cells as a function of storage time t at room temperature Electrolyte composition (1) PAN, LiClO4, PC (2) PAN, LiClO4, PC, 15C5

t=0

t = 9 months

σsp , S cm–1

σsp , S cm–1

2.1 × 10–4 5.0 × 10–4

2.8 × 10–4 2.3 × 10–4

RUSSIAN JOURNAL OF ELECTROCHEMISTRY

Vol. 37

We obtained and studied plasticized polymer electrolytes containing polyacrylonitrile, lithium salt, propylene carbonate, and crown ethers. The electrolytes have a high conductivity of about 10–3 S cm–1. We also investigated a new thin-film material, specifically, polyacetylene on porous polypropylene, which acts as a protective coating for a lithium anode. Systems Li/PE/Li are stable in a temperature range –10 to 50°C. Electrochemical characteristics of the Li/PE/Li system remain stable for a long storage time. ACKNOWLEDGMENTS The authors are grateful to V.M. Mazin for his help in performing electrochemical investigations and to L.S. Leonova, for her participation in discussing the results. REFERENCES 1. Yarmolenko, O.V., Ukshe, A.E., Yakushchenko, I.K., et al., Elektrokhimiya, 1996, vol. 32, p. 520. 2. Abraham, K.M. and Alamgir, M., J. Electrochem. Soc., 1990, vol. 137, p. 1657. 3. Alamgir, M. and Abraham, K.M., J. Power Sources, 1995, vol. 54, p. 40. 4. Zhu, Z.X., McMillan, R.S., and Murray, J.J., J. Electrochem. Soc., 1993, vol. 140, p. 101. 5. Belov, D.G., Efimov, O.N., and Belov, G.P., in Electrical and Optical Polymer Systems: Fundamentals, Methods, No. 3

2001

286

YARMOLENKO et al. and Applications, New York: Marcel Dekker, 1998, p. 297.

6. Tkachenko, L.I., Zueva, A.F., Saratovskikh, S.L., et al., Elektrokhimiya, 1992, vol. 28, p. 1818. 7. Efimov, O.N., Belov, D.G., Kozub, G.I., et al., Synth. Met., 1996, vol. 79, p. 193.

8. Gladyshev, G.P., Ershov, Yu.A., and Shustova, O.A., Stabilizatsiya termostoikikh polimerov (Stabilizing Thermostable Polymers), Moscow: Khimiya, 1979. 9. Ito, T., Shirakawa, H., and Ikeda, S., J. Polym. Sci., Polym. Chem. Ed., 1974, vol. 12, p. 11. 10. Ukshe, E.A. and Bukun, N.G., Tverdye elektrolity (The Solid Electrolytes), Moscow: Nauka, 1977.

RUSSIAN JOURNAL OF ELECTROCHEMISTRY

Vol. 37

No. 3

2001