Li-ion solvation in TFSI and FSI - based ionic liquid electrolytes probed by X-ray photoelectron spectroscopy

We demonstrate the capability of conventional laboratory XPS to determine the anions solvation shell of Li+ cation within 1M of LiTFSI and 1M of LiFSI salts dissolved in (EMIM+-FSI-) and (EMIM+-TFSI-) ionic liquids. The binding energy difference between the N1s components originating from the EMIM+ cation and from TFSI- or FSI- anions, solvating the Li+, confirms that both TFSI- and FSI- contribute simultaneously to the Li+ solvation. Additionally, the degradation of the TFSI and FSI -based electrolytes under X-ray exposure is proved.   

Structural formula of (a) EMIM+ cation, (b) TFSI- and (c) FSI- anions. N1s and S2p XPS core levels acquired on (a) 1M LiTFSI in [EMIM+-TFSI-] and (b) 1M LiFSI in [EMIM+-FSI-]. N1s acquired on (c) 1M LiFSI in [EMIM+-TFSI-] and (d) 1M LiTFSI in [EMIM+-FSI-]. The binding energy difference [ΔE=N1s(EMIM)-N1s(solvated Li)] between the N1s components originating from the EMIM+ cation and from TFSI- or FSI- anions allows to determine the solvation shell of the Li+ cation.

The continuous need for increasing the energy density, safety and electrochemical performance of Li-ion batteries (LiBs) requires an incessant development of characterization techniques capable of providing better insights into the physics and chemistry of the various parts in a battery. When explicitly investigating surface/interface modifications, X-ray photoelectron spectroscopy (XPS) is routinely employed. For LiBs, it is considered to be one of the suitable characterization techniques allowing to study the interface evolution occurring on cycled electrodes. The advantage of the XPS is the surface sensitivity providing direct information of e.g. the by-product species related to the electrolyte stability, in both liquid and solid -based electrolytes. XPS in LiBs is not limited to study the electrolyte/electrode interface chemistry, it can also be carried out to probe just the liquid electrolyte and investigate the Li-ion solvation for example in carbonate -based electrolytes as already demonstrated by our group using near ambient pressure photoemission (NAP) combined with a liquid jet. Such experiment and information are very precious, as the Li-ion solvation in the LiBs’ electrolyte play an important role in improving the cycling performance, by directly affecting the ionic conductivity, the stability at high or low potential and temperature, as well as the property of the solid electrolyte interphase (SEI). In this work, conventional XPS using Al Kα X-ray source is employed to investigate the cation and anion electronic structure and to monitor the Li-ion solvation within droplets of low vapor pressure and vacuum compatible ionic liquid electrolytes. Ionic liquid-based electrolytes are considered as promising alternative for next generation high energy density LiBs owing to their low vapor pressure and high thermal as well as electrochemical stability. However, not all ionic liquid-based electrolytes can successfully be deployed in LiBs. For example, stable electrochemical cycling of graphite in ionic liquid electrolytes can be achieved in presence of FSI- anion originating either from the LiFSI salt or as an anion from the ionic liquid. On the contrary, graphite cannot be cycled when TFSI- anion is used, unless FSI- is added in the electrolyte, despite the structural similarity between the anions. Intrigued about this phenomenon, we examined in this study the Li-ion solvation when both TFSI- and FSI- are present simultaneously in the electrolyte, as a salt or as anion in the ionic liquid. 1M of LiTFSI and 1M of LiFSI salts dissolved in (EMIM+ - FSI-) and (EMIM+ - TFSI-) ionic liquids respectively are investigated by acquiring the F1s, N1s, C1s, S2p and Li1s core levels. The abbreviation of EMIM, TFSI and FSI stand for 1-ethyl-3-methylimidazolium, bis(trifluormethanesulfonyl) imide and bis(fluorosulfonyl) imide respectively. The measurements of the binding energy difference [ΔE=N1s(EMIM)-N1s(solvated Li)] between the N1s component originating from the EMIM+ cation and the N1s component originating from TFSI- or FSI- anions, solvating the Li+, confirms that both TFSI- and FSI- contribute simultaneously to the Li+ solvation. This result can partially explain why adding FSI- in the electrolyte prevents the EMIM+ co-intercalation in the graphite by forming a favorable SEI during the first reduction. Moreover, this study sheds light on the stability of the ionic liquid electrolyte under the X-ray source, investigated by acquiring the F1s core level. It is worth mentioning that the measured pure ionic liquids EMIM+ TFSI- or EMIM+ FSI- without the salts show good stability under X-ray. No sign of any ionic liquid degradation by-product is detected and no change in the XPS core levels is observed even after long exposure time beyond 2 hours. However, as soon as the salts are added in the ionic liquid, then LiF component is detected as a by-product of the electrolyte degradation under the X-ray. This result limits our ambition in performing operando XPS measurements at the interface of ionic liquid electrolytes versus any working electrodes, where alternative solutions need to be invented for such an issue, like the use of liquid jet where fresh electrolyte is continuously measured.