Electrochemical Charge Transfer Reaction Kinetics at Silicon Electrode Surfaces

The volumetric lithiation capacity of silicon as an anode material is roughly ten times that of conventional graphite, while their densities are comparable. As a consequence, for a given particle size and C-rate, the current density required to cycle a silicon electrode at a given C-rate is about ten times greater than that of graphite. Such high current densities may cause the charge transfer kinetics at the silicon electrode surface to become rate limiting during use, depending on the magnitude of the corresponding Butler-Volmer exchange current density.  

Further motivated by the lack of consensus in the literature [1-4], we developed a novel approach to measure the exchange current density, using single crystal electronically conductive silicon wafers with well-defined crystallographic (100) orientation and active areas, and a cycling scheme to avoid stress-induced surface cracking and fragmentation upon lithiation. Electron microscopy, cyclic voltammetry, and electrochemical impedance spectroscopy were used as characterization tools. It was demonstrated that the silicon electrode surface remains crack free during the first lithiation half cycle, with fracture being induced only during the following delithiation half cycle, which is in agreement with the observations made by Chon et al. [5]. Consequently, it was possible to grow the solid electrolyte interface (SEI) layer as is characteristic of silicon powder electrodes, while still using the geometric area as the electrochemically active surface area to evaluate the surface kinetic parameters from EIS experiments.

The exchange current density at the silicon electrode surface was found to be 0.1 ± 0.01 mA/cm² for electrolyte consisting of 1 M LiPF₆ in EC/DMC (1/1)w + FEC (10%)w + VC (2%)w. Figure 1 shows the corresponding kinetic overpotential as a function of particle diameter and C-rate for silicon and graphite. Clearly, the overpotential is much greater in the silicon case. The impact of these results on cell performance is discussed. The measured exchange current density values may be used as input for porous electrode models, and to devise electrode microstructures and cycling regimes for improved fracture tolerance and overall performance.

Acknowledgments

This work was supported as part of the NorthEast Center for Chemical Energy Storage (NECCES), an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences under Award # DE-SC0012583.

References

[1] L. Baggetto et al., J. Power Sources, 189, 402–410 (2009)

[2] L. Baggetto, R. A. H. Niessen, and P. H. L. Notten, Electrochim. Acta, 54, 5937–5941 (2009)

[3] V. A. Sethuraman, V. Srinivasan, and J. Newman, J. Electrochem. Soc., 160, A394–A403 (2013)

[4] R. Chandrasekaran, A. Magasinski, G. Yushin, and T. F. Fuller, J. Electrochem. Soc., 157, A1139 (2010)

[5] M. J. Chon, V. A. Sethuraman, a. McCormick, V. Srinivasan, and P. R. Guduru, Phys. Rev. Lett., 107, 045503 (2011)