Determination of the Solid Electrolyte Interphase Structure Grown on a Silicon Electrode Using a Fluoroethylene Carbonate Additive

Abstract

In this work we explore how an electrolyte additive (fluorinated ethylene carbonate – FEC) mediates the thickness and composition of the solid electrolyte interphase formed over a silicon anode in situ as a function of state-of-charge and cycle. We show the FEC condenses on the surface at open circuit voltage then is reduced to C-O containing polymeric species around 0.9 V (vs. Li/Li+). The resulting film is about 50 Å thick. Upon lithiation the SEI thickens to 70 Å and becomes more organic-like. With delithiation the SEI thins by 13 Å and becomes more inorganic in nature, consistent with the formation of LiF. This thickening/thinning is reversible with cycling and shows the SEI is a dynamic structure. We compare the SEI chemistry and thickness to 280 Å thick SEI layers produced without FEC and provide a mechanism for SEI formation using FEC additives.

Lithium insertion into silicon electrodes studied by cyclic voltammetry andoperandoneutron reflectometry

“…A discrepancy is also found in the SLD of the SEI that is ranging between 2 Â 10 À6 Å À2 and 4 Â 10 À6 Å À2 in the work of Veith et al 22 compared to values found here of 0.4 Â 10 À6 Å À2 to 1.2 Â 10 À6 Å À2 . In a different study 24 of the same group, where fluorinated ethylene carbonate was used as an additive to the electrolyte, also a thickening of the SEI was found during lithiation by some tenths of a nanometre and a reduction during delithiation, in qualitative agreement with our work. Fears et al 23 found by NR and X-ray photoelectron spectroscopy (XPS) also SEI growth and shrinkage during electrochemical cycling.…”

“…Before further analysis of the experimental data is done, the following point has to be stated: we assumed in our model a spatial constant Li concentration x within the a-Li x Si layer (homogenous Li distribution) where x increases with lithiation time and decrease with delithiation time as also done by other authors. 22,24 This model assumption is not a priori valid. Also a spatial heterogeneous Li distribution in the a-Li x Si layer might be present at least during initial stages of the first cycle during galvanostatic cycling.…”

“…22 and 23. Veith et al 24 investigated also via in situ NR, X-ray photoelectron spectroscopy and IR-spectroscopy the influence of FEC (fluorinated ethylene carbonate) additives in the electrolyte (1.2 M LiPF 6 -ethylene carbonate:dimethyl carbonate (EC : DMC 3 : 7 wt%)) on the formation and composition of the SEI at different states of charge. They observed even under open circuit conditions the formation of a 50 Å thin SEI which thickens and becomes more organic during lithiation.…”

New insights into the solid–liquid interface exploiting neutron reflectivity

“…[13] An alternative approach to study surfaces which are not transparent to neutrons, is to support a thin film of the material on a transparent substrate. This has been reported for an increasing number of metal systems including Fe [14,15], Ti [16][17][18], Cu [19][20][21][22][23][24][25], Ni [26][27][28], Al [29,30] and Au [31][32][33][34]. Steel [35] remains a challenge with the overall deposited composition replicated, but the metallurgical aspects are not necessarily typical of bulk steel (see section 2.3 below).…”

“…NR has been used to identify a number of characteristics of the SEI including irreversible Li incorporation, [25] formation of a "skin",[57] the report of a "breathing" mode [24] and the important role of additives [23], with precise details depending on the electrode and electrolyte used. Importantly there are discrepancies between these in-situ NR measurements and complementary ex-situ XPS data.…”