No abstract

Due to the flammability of liquid electrolytes used in lithium ion batteries, solid lithium ion conductors are of interest to reduce danger and increase safety. The two dominating general classes of electrolytes under exploration as alternatives are ceramic and polymer electrolytes. Our group has been exploring the preparation of molecular solvates of lithium salts as alternatives. Dissolution of LiCl or LiPF in pyridine (py) or vinylpyridine (VnPy) and slow vapor diffusion with diethyl ether gives solvates of the lithium salts coordinated by pyridine ligands. For LiPF, the solvates formed in pyridine and vinylpyridine, namely tetrakis(pyridine-κN)lithium(I) hexafluorophosphate, [Li(CHN)]PF, and tetrakis(4-ethenylpyridine-κN)lithium(I) hexafluorophosphate, [Li(CHN)]PF, exhibit analogous structures involving tetracoordinated lithium ions with neighboring PF anions in the I-4 and Aea2 space groups, respectively. For LiCl solvates, two very different structures form. catena-Poly[[(pyridine-κN)lithium]-μ-chlorido], [LiCl(CHN)], crystalizes in the P222 space group and contains channels of edge-fused LiCl rhombs templated by rows of π-stacked pyridine ligands, while the structure of the LiCl-VnPy solvate, namely di-μ-chlorido-bis[bis(4-ethenylpyridine-κN)lithium], [LiCl(CHN)], is described in the P2/n space group as dinuclear (VnPy)Li(μ-Cl)Li(VnPy) units packed with neighbors via a dense array of π-π interactions.

The motivation for the development of solid electrolytes arises from safety issues associated with liquids electrolytes currently being used in commercial lithium ion batteries. Solid electrolytes made from ceramic/glass possess highest ionic conductivities, in the range of 10-3 to 10-2 S/cm [1], but are brittle, with poor adhesion to the electrodes and are difficult to process. Solid polymer electrolytes such as polyethylene oxide (PEO) [2], PEO/composite blends, PEO copolymers/blends [3-4], molecular or ionic plastic crystals, and low molecular weight glymes have lower ambient temperature conductivities (10-7-10-5S/cm), but are flexible, with better adhesion to electrodes, and are easier to process. The low ionic conductivity of aliphatic oxide solid electrolytes (for example: PEO) is due to the significant affinity of Li+ ions for ether oxygen atoms, the large number of O-Li contacts/Li,  and the coupling of ion migration to slow back bone segmental motions. Based upon the Pearson Hard-Soft Acid-Base (HSAB) concept [4], hard (charge dense, non-polarizable) lithium cations have high affinity for hard ether oxygen atoms, resulting in low Li+ mobility and thus low ionic conductivity. Various approaches have been attempted to improve overall ionic conductivity/cation mobility such as molecular organization of ordered rather than disordered structures and decreased interactions between Li+mobile ions and their counter anions/solvating matrix[5-9].   We have adopted another approach to the formation of solid electrolytes, referred to here as soft-solid electrolytes, in which lithium salts are trapped in atomic size organic channels, where there is low affinity of the channel walls for the ions, allowing higher ionic conductivity under an applied field. In the current work, we have prepared crystalline, solid electrolytes with organic matrices by cocrystallization of soft (charge diffuse, polarizable) N,N-dimethylformamide (DMF) donors and LiCl. The soft C=O functionality in DMF interacts poorly with the hard Li+ ions, based on HSAB. Cocrystallization of these molecules results in low-affinity ion channels of lithium and superior conductivity properties for an organic solid-state electrolyte. Single crystal X-ray data identifies the solid as 1:1 adduct of DMF:LiCl (Figure 1). The structure shows the formation of 1-D ionically bonded Li2Cl2 rhombs interacting with the soft DMF matrix through weak Lewis acid/base interactions of lithium and oxygen ions. The DMF•LiCl crystal exhibits a linear arrangement of alternating Li-Cl and Li-O(DMF) rhombs arranged end to end through lithium atoms, such that a chain of closely spaced (2.9 Å) lithium ions exists parallel to the aaxis of the unit cell (Figure 2). Pressed pellets of DMF:LiCl exhibit excellent room temperature ionic conductivity 1.3x10-4 S/cm (Figure 3), a lithium ion transference number (tLi +) of 0.25 and reasonable interfacial properties with lithium metal. The pressed pellet exhibits both bulk and grain boundary resistance, with a liquid-like interfacial layer of LiCl-DMF possible providing contact between the grains. (1) Kamaya, N.; Homma, K.; Yamakawa, Y.; Hirayama, M.;  Kanno, R.; Yonemura, M.; Kamiyama, T.; Kato, Y.; Hama, S.; Kawamoto, K.; Mitsui, A. Nature Materials 2011, 10, 682. (2) P.R.Chinnam, S.L.Wunder. Journal of Materials Chemistry A 2013, 1, 1731 (3) P.R.Chinnam, Zhang, H, S.L.Wunder. Electrochimica Acta 2015 (doi:10.1016/j.electacta.2015.04.010) In press (4) P.R.Chinnam, S.L.Wunder. Chemistry of Materials 2011, 2011, 23, 5111. (5) Pearson, R. G. Journal of the American Chemical Society 1963, 85, 3533. (6) Borodin, O.; Smith, G. D. Macromolecules 2006, 39, 1620. (7) Bruce, P. G. Philosophical Transactions of the Royal Society a - Mathematical Physical and Engineering Sciences 1996, 354, 1577. (8) Andreev, Y. G.; Bruce, P. G. Electrochimica Acta 2000, 45, 1417. (9) Golodnitsky, D.; Peled, E. Electrochimica Acta 2000, 45, 1431. Figure 1