All Solid-State Electrolytes for Lithium Microbatteries: Fabrication of an Ionic Liquid-Based Membrane

Through the permanent miniaturization of portable autonomous systems, the development of reliable, integrated micro power sources is necessary. Currently, Li-ion batteries are limited by their size, which make them unsuitable to new technologies on the millimeter scale or less. In this context, lithium-based microbatteries raise great interest at both academic and industrial levels. These systems are less than 10µm thick and a few square centimeters wide with storage capacities going from 50 to 200µAh/cm². Their convenient size make them an effective solution for applications such as self-powered microelectronics (connected smart-cards, ultra-thin watches, autonomous sensors, etc…), medical devices (pacemaker, hearing aid, etc…) and energy harvesting devices (solar cells). However, microbatteries are built by successive thin films deposition, mainly by PVD processes, which involve expensive technologies and time-costly processes. In order to reduce manufacturing costs and also enable conformal deposition on 3D substrates to increase energy storage capabilities, our laboratory is currently developing milder deposition routes for microbatteries films. For instance, standard electrolyte, LiPON, is a glassy thin film deposited at a rate of 0.3µm/hour . This project focuses on the study of new solid electrolytes for microbatteries through an innovative wet process. Our strategy consists in confining a mixture of an ionic liquid and a lithium salt in an organic membrane in-situ polymerized. Ionic liquids (IL) are interesting electrolyte solvents because of their high ionic conductivity, paired with their excellent chemical and electrochemical stability vs Li + /Li. Moreover, IL have important features in terms of security and process flow, due to their low flammability and thermal stability up to 350°C. A wide range of anion-cation couples can tailor the material properties. Thus, we studied the characteristics of several ionic liquid/lithium salts binary solutions. Three different cations families, 1-Ethyl-3-Methylimidazolium (EMI), N-alkykl-N-methylpyrrolidinium (Pyr 1A + ) and 1-Alkyl-1-Methylpiperidinium (PIP 1A + ), were combined to two anions, bis(trifluoromethane-sulfonyl)imide (TFSI - ) and bis(fluorosulfonyl)imide (FSI - ). Electrochemical stability window, ionic conductivity and ionic transport were investigated as well as temperature stability. The chosen electrolytic solution was then confined in an organic matrix by photo-induced (UV light) radical polymerization of a di-methacrylate precursor. The reticulation was tracked by differential scanning calorimetry associated with a photo-calorimetric accessory (PCA-DSC) and by infrared spectroscopy (FTIR). Additional interesting results were obtained by following the conductivity evolution during the polymerization. For this purpose, we designed a completely original planar interdigitated cell to measure reliable electrochemical impedance during UV-curing. The influence of curing parameters such as precursor concentration and light intensity on the final gel conversion and structure were also investigated. After optimal polymerization, thermally stable (up to 300°C), transparent and flexible gelled electrolytes called “ion-gels” were obtained – Figure 1 . Synthesized ion-gels displayed ionic conductivity around 10 -4 S.cm -1 at 25°C, significantly better than the standard electrolyte (LiPON) conductivity of 10 -6 S.cm -1 . Our electrolytes also had broad stability window vs Li + /Li up to 5V. Indeed, electrochemical behavior of ionic liquids is almost completely preserved after their confinement in the matrix. A Li/ion-gel/LiCoO 2 coin cell was assembled and characterized by impedance spectroscopy (EIS) and galvanostatic cycling (GCPL) – Figure 2 . Since the electrolyte precursor solution is liquid before UV-curing, a thin film with controlled thickness (50-200µm) can be easily deposited on the positive electrode. The battery has successfully cycled at different rates (C/10, C/5, C/2, C = 30µA/cm²) between 3 and 4.2V at 25°C. Its total capacity was equivalent to 80% of the “reference” capacity obtained with a common carbonate-based liquid electrolyte (EC-PC-DMC + 1M LiPF 6 ). The capacity loss has been assigned to the polarization, which increases with the cycling rate, caused by pristine rough interfaces and consequently higher electrolyte resistivity. However, first results showed that the interfaces and the whole electrolyte film are stable for hundreds of cycles. A functional Li-microbattery Li/ion-gel film/LiCoO 2 prototype was also produced. The lithium layer was deposited on the ion-gel by evaporation deposition without degradation of the electrolyte during the process. Figure 1