Nucleation, Growth, and Properties of the Solid Electrolyte Interphase – a Multimodal Approach Using a Model System

Due to the low potentials on anode materials in lithium-ion batteries, the electrolyte decomposes on the anode forming a passivation layer, termed solid electrolyte interphase (SEI). The SEI plays an essential role in battery performance since it is a major source of capacity loss and dictates cell kinetics. Nevertheless, the properties of the SEI are not yet well-understood.

Here, we present the utilization of a simple and well-defined model system to understand the atomic level SEI nucleation and growth, as well as structural and chemical properties. Experiments were performed in a half-cell configuration with N-doped silicon carbide (N-SiC) wafer working electrode, Li metal counter/reference electrode, and LP30 electrolyte (1M LiPF₆ in 1:1 wt% ethylene carbonate:dimethyl carbonate). N-SiC has the advantage of being electrically conductive but electrochemically inactive, allowing for a straightforward interpretation of electrochemical and structural measurements, and provides information on the intrinsic electrolyte reactivity.

We used a multimodal approach combining in-situ X-ray reflectivity (XRR), precision electrochemistry at slow rates (e.g. 0.16 mV/s cyclic voltammetry), X-ray photoelectron spectroscopy (XPS), and X-ray diffraction (XRD). We obtained the evolution of the SEI's thickness, electron density, and roughness through quantitative operando XRR measurements during the first two cyclic voltammetry (CV) cycles. Further, we found via complementary XPS experiment that the SEI was mainly composed of LiF, and we tracked the crystalline LiF growth with in-situ XRD. LiF can be formed via two reaction pathways. One is the direct reduction of PF₆^-, the other is the electrocatalytic hydrogen evolution reaction of the trace amount of HF in the electrolyte, forming LiF and H₂. In order to elucidate the formation mechanism of LiF, we combined the electrochemical and structural/chemical results for LP30 with precise electrochemistry using several baseline LiPF₆ and/or HF containing electrolytes. The results indicate that the direct reduction of PF₆^- contributes to LiF formed at ~ 1.5 V versus Li/Li⁺, consistent with our quantum chemistry calculations. Finally, we revealed that the ratio between organic and inorganic compounds in SEI could be tuned by changing the cycling rate.

We envision that this methodology and the use of well-defined model systems can be applied to other electrolyte-electrode systems. Our results are essential in understanding the formation mechanism and properties of SEI and can aid future experimental and computational works.