Elsevier

Journal of Power Sources

Volume 135, Issues 1–2, 3 September 2004, Pages 232-239
Journal of Power Sources

Crosslinkable fumed silica-based nanocomposite electrolytes: role of methacrylate monomer in formation of crosslinked silica network

https://doi.org/10.1016/j.jpowsour.2004.03.064Get rights and content

Abstract

The electrochemical and rheological properties of composite polymer electrolytes (CPEs) based on fumed silica with tethered crosslinkable groups are reported. These silica nanoparticles are dispersed in electrolytes consisting of poly(ethylene glycol) dimethyl ether (PEGdm)+lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) to which various methacrylate monomers, such as methyl (MMA), ethyl (EMA), butyl (BMA), n-hexyl (HMA), and n-dodecyl (DMA) methacrylate, are added. The methacrylate monomer facilitates creation of chemical crosslinks between fumed silica particles and formation of a crosslinked network. In this study, the effects of concentration and alkyl chain length of the monomers on conductivity, dynamic rheology, open-circuit interfacial stability, and cell voltage in lithium–lithium cell cycling are examined. Increasing the length of the monomer alkyl chain enhances both conductivity and elastic modulus of the crosslinked CPE. In contrast, increasing monomer concentration results in higher elastic modulus, but reduced conductivity. Lithium–lithium cell cycling and open-circuit interfacial stability results did not correlate with alkyl chain length. That is, for the lithium–lithium cycling studies, all crosslinked samples exhibit higher half-cycle voltage compared to non-crosslinked samples; however, the open-circuit interfacial stability of CPEs containing BMA and HMA exhibit improved stability compared to the other monomers and the CPE without monomer.

Introduction

Rechargeable lithium batteries are promising power sources for devices such as electric vehicles, portable electronics, and implantable medical devices, because of their high-energy density and low self-discharge rate. Although significant progress has been made in these batteries, several important factors, especially the low conductivity and chemical stability of the electrolyte, have limited their commercial use. A large portion of electrolyte research focuses on improving the mechanical strength of the electrolyte without sacrificing important electrochemical properties, such as conductivity, lithium transference number, and interfacial stability.

Electrolytes for lithium batteries must have acceptable ionic conductivity (>10−3 S cm−1 at 25 °C) and should possess a high Li+ transference number, i.e., a high ratio of the charge transported by Li+ compared to the total charge transported [1], [2], [3]. Electrolytes must be chemically and electrochemically stable, mechanically strong, yet easily processable, safe, and inexpensive [1], [2], [3], [4], [5], [6]. Solid polymer electrolytes have been recognized as viable candidates for rechargeable lithium batteries. The use of solid polymer electrolytes would overcome limitations of liquid electrolytes including: lithium dendrite formation, electrolyte leakage, flammable organic solvent, and electrolytic degradation of electrolyte [5], [7], [8].

Early approaches to solid polymer electrolytes employed high-molecular weight (Mn>105) polyethylene oxide (PEO) [9], [10], [11]. Linear PEO forms a semi-crystalline electrolyte with reasonable mechanical properties after addition of lithium salts. The crystallites act as crosslinks that make high-molecular PEO dimensionally stable up to the melting point of the salt–polymer complexes (∼65 °C). However, since conductivity is dominated by the segmental motion of the amorphous polymer, the presence of crystalline phase reduces the room-temperature conductivities to less than 10−5 S cm−1, which is too low for most practical applications [10], [11]. Various methods have been devised to inhibit crystallinity and maximize the volume fraction of the amorphous phase in PEO, including: (a) synthesis of PEO-based polymers with modified architectures [1], [12], [13], [14], [15]; (b) addition of ‘plasticizers’ [10], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25]; (c) addition of solvents [6], [24], [26], [27], [28], [29], [30], [31], [32], [33], [34]; (d) addition of inorganic fillers such as alumina, silicas, and zeolites [11], [14], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48].

The synthesis of PEO-based polymers with modified architectures has emphasized comb-branched systems, where the “teeth” of the comb are designed to have maximum flexibility and hence high conductivity. However, the mechanical properties of these polymers are poor and require crosslinking to minimize creep. Even for these neat polymer systems, regardless of the polymer structure, the room-temperature conductivity [15] of electrolytes is less than 5×10−5 S cm−1. The addition of plasticizers to PEO, either as an absorbed liquid or as a plasticizing lithium salt [10], [16], [23], [24], [25], is often used to reduce the glass transition temperature Tg and increase conductivity to 5×10−4 S cm−1; however, such additives often cause a reduction in mechanical properties and a decrease in the interfacial stability with lithium [2], [11], [49].

Gel electrolytes are two-component systems that are prepared by dispersing liquid solvents (typically, organic carbonates) and/or plasticizers in an electrochemically inert polymer, such as polyacrylonitrile (PAN) and poly(vinylidene fluoride) (PVdF) [12], [13], [50], [51], [52], [53]. Here, the ionic transport is primarily governed by the liquid electrolyte with the polymer providing mechanical support. The ionic conductivity of a gel electrolyte is higher than that of a solid system, but often at the expense mechanical strength. To improve the mechanical properties of the gel, crosslinkable components (e.g., diacrylate compounds) may be added to form networked gel–polymer systems [51], [52].

The addition of inorganic fillers to form composite polymer electrolytes (CPEs) decreases the crystallinity in samples prepared from high-molecular weight PEO, thus stabilizing the conductive amorphous phase. Furthermore, the fillers can form self-assembled network structures that provide favorable mechanical properties in low-to-moderate molecular weight electrolyte systems [54], [55], [56], [57], [58], [59], [60], [61], [62], [63]. The principal advantage of the self-assembly approach is that it provides significant processing advantages that can lead to reduced-cost electrolytes.

CPEs consisting of low-molecular weight polyethers, lithium salts, and fumed silica are being developed in our laboratories to produce materials with high conductivity and mechanical stability [54], [55], [60], [63]. The mechanical stability stems from a three-dimensional network of interacting fumed silica aggregates. A unique feature of our CPEs is that the surface chemistry of the fumed silica particles can be tailored to produce desirable mechanical properties without affecting the electrochemical properties. We have investigated composite electrolytes that contain fumed silica with octyl and methacrylate groups on the surface. Composites that contain these dual-functionalized fumed silica can be subsequently reacted to form a chemically crosslinked CPE rather than a physically crosslinked CPE [64]. Crosslinked CPEs require the use of additional monomers to form sufficient links between particles to give mechanically robust networks. Ideally, added monomer “coats” the surface of the silica network and provides permanent mechanical stability with minimal penalty in conductivity. Using commercially available crosslinkable fumed silica (DegussaR711)+PEGdm(250)+lithium bis(trifluoromethanesulfonyl)imide (LiTFSI, Li:O=1:20) + 20 wt.% butyl methacrylate (BMA) monomer, we prepared thermally crosslinked composites with a room-temperature ionic conductivity approaching 10−3 S cm−1 and an elastic modulus [56] (G′) approximately 105 Pa. We have recently built on this preliminary study to examine the effects of fumed silica surface group, fumed silica weight percent, salt concentration, and solvent molecular weight on electrochemical and rheological properties [65]. These composite electrolytes of crosslinked fumed silica exhibit significantly higher elastic modulus and yield stress with minimal penalty in conductivity and ion transport compared to the CPEs with hydrophobic fumed silica. However, these crosslinked systems behaved poorly in lithium–lithium cycling experiments due to the possible presence of unreacted monomer and/or poor interfacial contact.

In this study, we focus on understanding how changes in the monomer bulk properties affect the rheological and electrochemical properties of CPEs using methacrylate monomers of various length and concentration. Results from ionic transport, dynamic rheology, open-circuit interfacial stability, and lithium/lithium cell-cycling are discussed in terms of a mechanistic crosslinking model. In addition, these results are correlated with the bulk properties of the material components to further evaluate the performance of the composites, especially the interfacial stability with lithium.

Section snippets

Experimental

Chemicals and preparation used in this study consist of five components: crosslinkable-based fumed silica (Degussa R711), lithium salt (lithium bis(trifluoromethanesulfonyl)imide [LiN(CF3SO2)2] (LiTFSI, Li imide, 3M)), poly(ethylene glycol)dimethyl ether (PEGdm, Mn=250, Aldrich), methacrylate monomer, and initiator [55], [66]. LiTFSI is dried at 110 °C under vacuum (∼1 kPa) for 24 h before use. Fumed silicas are dried at 90 °C under vacuum (∼1 kPa) for 1 week to achieve a water content of 150–200 ppm

Effect of monomer weight percent

Crosslinkable fumed silica-based composite polymer electrolytes require the addition of monomer to tether adjacent fumed silica particles together and form a robust solid electrolyte. The addition of monomer to the composite electrolyte can significantly affect the electrochemical and rheological properties. To understand how the amount of monomer influences conductivity, we measured conductivity as a function of monomer weight percent. Fig. 1 shows the conductivity of crosslinked CPEs

Summary and conclusions

CPEs containing crosslinkable fumed silica using different methacrylate monomers have been prepared and characterized. The conductivity of the crosslinked CPEs increases as the length of the alkyl chain of the monomer increases with DMA>HMA>BMA>EMA>MMA. The rheological properties of the crosslinked CPEs also depend on the length of the alkyl chain. Longer alkyl chain monomers, such as BMA, HMA, and DMA, have higher elastic moduli than short alkyl chain monomers, such as MMA and EMA. Increases

Acknowledgements

The authors gratefully acknowledge funding from the Department of Energy, Office of Basic Energy Sciences and Office of FreedomCAR Vehicle Technologies. In addition, we acknowledge the US Department of Education Graduate Assistance in Areas of National Need (GAANN) Fellowship for providing additional funding. We also thank Degussa for providing the fumed silica and 3M for providing the LiTFSI salt.

References (70)

  • D. Fauteux et al.

    Electrochim. Acta

    (1995)
  • K.L. Heitner

    J. Power Sour.

    (2000)
  • Z.C. Zhang et al.

    Electrochim. Acta

    (2000)
  • M. Armand

    Solid State Ionics

    (1994)
  • J.Y. Song et al.

    J. Power Sour.

    (1999)
  • W. Wieczorek et al.

    Electrochim. Acta

    (1995)
  • O. Buriez et al.

    J. Power Sour.

    (2000)
  • Y.T. Kim et al.

    Solid State Ionics

    (2002)
  • R. Huq et al.

    Electrochim. Acta

    (1992)
  • S. Izuchi et al.

    J. Power Sour.

    (1997)
  • M.C. Borghini et al.

    Electrochim. Acta

    (1996)
  • Y.K. Kang et al.

    J. Power Sour.

    (2001)
  • S.I. Moon et al.

    J. Power Sour.

    (2000)
  • A. Noda et al.

    Electrochim. Acta

    (2000)
  • A. Zalewska et al.

    Solid State Ionics

    (2003)
  • C.S. Kim et al.

    Electrochim. Acta

    (2000)
  • K.M. Abraham et al.

    J. Power Sour.

    (1993)
  • S.R. Starkey et al.

    Electrochim. Acta

    (1997)
  • B.Y. Huang et al.

    Solid State Ionics

    (1996)
  • G.B. Appetecchi et al.

    Electrochim. Acta

    (1995)
  • J.E. Weston et al.

    Solid State Ionics

    (1982)
  • W. Gang et al.

    Solid State Ionics

    (1992)
  • W. Wieczorek

    Mater. Sci. Eng. B

    (1992)
  • F. Croce et al.

    Solid State Ionics

    (1990)
  • W. Wieczorek et al.

    Solid State Ionics

    (1989)
  • B. Kumar et al.

    Solid State Ionics

    (1999)
  • G.B. Appetecchi et al.

    Electrochem. Commun.

    (1999)
  • M. Kono et al.

    Electrochim. Acta

    (2000)
  • J. Fan et al.

    Solid State Ionics

    (1998)
  • H.J. Walls et al.

    J. Power Sour.

    (2000)
  • S.R. Raghavan et al.

    J. Colloid Interface Sci.

    (1997)
  • W.H. Meyer

    Adv. Mater.

    (1998)
  • R.J. Brodd et al.

    Macromol. Symp.

    (2000)
  • P.G. Bruce et al.

    J. Chem. Soc., Faraday Trans.

    (1993)
  • H.J. Gores et al.

    Pure Appl. Chem.

    (1995)
  • Cited by (9)

    View all citing articles on Scopus
    View full text