Solid-state lithium ion electrolytes have the potential to improve the safety and reliability of lithium batteries, but they are typically several orders of magnitude lower in ionic conductivity than their liquid counterparts. One way to address this problem is to use thin film electrolytes.  By reducing the thickness of the electrolyte, the distance that Li ions must travel is decreased which helps reduce the negative effects of a lower ionic conductivity.  Recently the crystalline solid electrolyte Li₁₀GeP₂S₁₂ (LGPS) achieved Li⁺ conductivities of up to 1.2x10^-2 Scm^-1, which is approximately equivalent to liquid electrolytes [1]. This increase in conductivity is attributed to site-to-site hopping of Li ions in the tetrahedral LGPS solid [2]. Other LGPS-type electrolytes, like Li₁₁Si₂PS₁₂ have also exhibited fast Li ion hopping through tetrahedra in the bulk of the material with Li ion diffusion coefficient up to 3.5x10^-12 m²s^-1, similar to LGPS’s 2.2x10^-12 m²s^-1[3]. Previous studies have been conducted in crystalline and bulk glass materials with tetrahedral backbones. To achieve thin films with necessary electrolyte performance, we chose radio frequency magnetron sputtering.  This allowed us to grow high quality thin films with compositional flexibility.  Sputtered thin films are likely to have amorphous character, so we investigated how amorphous thin films analogous to high performing materials, but using less expensive and rare components, affect the potential for fast Li ion hopping through composition and structure analysis. 

The desired thin films were produced through co-deposition from independently controlled targets onto substrates held on a rotating platform.  Varying RF powers were placed on a Li₂O target and a SiO₂target in order to create a range of stoichiometric lithium silicate electrolytes. These thin films were deposited on silicon or KCl chips.  Deposition duration was controlled to grow films to the requisite thickness for each composition and analysis technique. The Li/Si ratio of each thin film was determined by in inductively coupled plasma optical emission spectroscopy (ICP-OES). Samples were prepared for this analysis by dissolving a thin film deposited onto a KCl chip in DI water.  Further compositional and structural analysis was completed using x-ray photoelectron spectroscopy (XPS).  Film thickness was determined through profilometry. 

In studies with lithium silicate glasses, local depolymerization of SiO₄ tetrahedra created nanoscale phase separation, or Li-rich regions embedded in an insulating Si rich matrix. It has been shown that increased SiO₂ content in glasses constrict Li percolation paths between these Li-rich regions thus strongly influencing Li mobility [4].  The structural arrangement of bonding oxygen (BO) and non-bonding oxygen (NBO) per SiO₄^4-tetrahedral unit will be defined for each composition of thin film through raman spectroscopy. These local structural arrangements will help characterize the degree of nanoscale phase separation in our thin film.  Phase purity in the higher scale structural bulk of our material will be verified though scanning electron microscopy with energy dispersive x-ray spectroscopy (SEM EDS).  The cluster-tissue model proposes that glasses cooled at a high rate form clustered pseuodophases surrounded by regions of relatively more amorphous connective tissue [5].  We used Infrared Spectroscopy to examine if this structure is observed in our sputtered thin films, which essentially are glasses with even higher cooling rates. 

A detailed understanding of the structure of our lithium silicate thin films will allow us to understand the framework through which Li ions must travel through this amorphous electrolyte.  Future studies will include using impedance spectroscopy to analyze how the ionic conductivities are affected by the structural differences between each composition that we have defined in this work.

1. Kamaya, N. et al. Nat Mater 10, 682–686 (2011).
2. Kuhn, A., Duppel, V. & Lotsch, B. V. Energy & Environmental Science 6, 3548 (2013).
3. Kuhn, A. et al. Phys. Chem. Chem. Phys. 16, 14669–14674 (2014).

4. Bauer, U. et al. J. Phys. Chem. B 117,15184–15195 (2013).

5. Goodman, C. H. L. Nature 257, 370–372 (1975).