Interfacial analysis is critical to analyze the performance of Li-ion batteries, the prediction of reactions occurring at the interface electrode/electrolyte, the formation of the solid electrolyte interface (SEI), affecting the cycling life of the battery. These are issues that computational techniques can address prior to experiments and fabrication. We analyze interphases produced by liquid-solid and solid-solid electrolyte-anode interfaces using molecular dynamics simulations (MD);^1-4 thus, experience from liquid electrolytes SEIs is extended to the analysis of alternative solid electrolytes and their SEIs with the anode.

The solid electrolyte interphase formed due to a liquid-solid electrolyte-anode is modeled entirely as a LiF layer coating a LiSi nanoparticle (NP). We focus on the mechanical properties of the SEI layer during the expansion of the NP due to charging of a Li-ion battery.² Several charging rates are tested presenting current dependent phenomena such as a surface reconstruction, amorphization and cracking of the SEI. Extrapolating C-rates, we identified an interval between cracking onset and critical cracking (SEI fully fragmented); this interval corresponds to a volume expansion from 180 to 320%, matching the experimental range of charging maximum of 2200 mAh/g and equivalent to ~200% volume expansion.⁵ The breaking of LiF bonds begins with those nearest to the SiLi-LiF interface, along a radial direction pointing to the outside of the NP, then forming larger and larger LiF rings until the SiLi alloy totally loses its protective shell. Due to cracking, the anode is exposed to the electrolyte allowing the formation of dendrites. Taking a sample of a cracked LiF, a MD simulation is performed to develop and test a protocol to show the formation of dendrites and what influences them, making them faster or slower to grow.¹ In order to avoid, or at least reduce the effect, of the dendrites, we test solids materials as electrolyte.

For the solid-solid interfaces, we analyze the ClN and Li (001) planes of the solid Li₉N₂Cl₃ electrolyte interfacing the Li-metal (001) using ab initio MD to identify the electrochemical reactions that may occur at the formation of the interface of an initially discharged battery with or without out a charging power source.⁶ An atomic concentration analysis does not reveal any significant changes up to 20 ps in any of the two analyzed interfaces, meaning that there is no considerable migration of atoms from electrode to electrolyte and vice versa. Diffusion coefficient of all the atomic species involved decreases with a trend to reach a minimum value, indicating that the atoms reach a stable state, trying to avoid movement. Charge transfer on the Li sites at the interphases is observed, changing its charge from 0 to +1. On interface solid-ClN—Li-metal, N and Cl atoms form bonds with 7-8 Li around them, taking Li from the electrode to achieve a BCC structure with N and Cl as central atoms. On the solid-Li—Li-metal, one Li, originally from the electrolyte, diffuses to the electrode, however, this phenomenon has only occurred once during the whole simulation, indicating that it is possible but it is not recurrent; so at least during the equilibration, the loss of Li of the electrolyte to the electrode is not a considerable problem in these interfaces. However, other solid electrolytes react at the interface forming compounds. For example, Li₅PS₄Cl, Li₇P₂S₈Br, and Li₇P₂S₈I electrolytes, derivatives such as Li₂S, Li₃P and LiCl form in the Li/LPSC interface at 50 ps on each face of the solid electrolyte and from analyzing the faces of LPSC, the highest Li-metal consumption is on the (010) face.

Author Contributions: D.E.G.-A analyzed the mechanical properties of the LiF solid electrolyte interface and studied the Li/Li9Na2Cl3 interface; L.A. S. performed the molecular dynamics simulation of dendrite formation; C.R.-V. did the interfacial study of Li/Li5PS4Cl2; F.F.-G. did the study of the interfacial Li/Li7P2S8I; M.G. did the study of the interfacial Li/Li7P2S8Br; and J.M.S. directed the research.

1. Selis, L. A.; Seminario, J., RSC Advances 2018, 8 (10), 5255-5267.
2. Galvez-Aranda, D. E.; Seminario, J. M., J. Electrochem. Soc. 2018, 165 (3), A717-A730.
3. Galvez-Aranda, D. E.; Ponce, V.; Seminario, J. M., J. Mol. Model. 2017, 23 (4), 120.
4. Ponce, V.; Galvez-Aranda, D. E.; Seminario, J. M., J. Phys. Chem. C 2017, 121 (23), 12959-12971.
5. Yang, Y.; Wang, Z.; Zhou, R.; Guo, H.; Li, X., Materials Letters 2016, 184, 65-68.
6. Roman-Vicharra, C.; Franco-Gallo, F.; Alaminsky, R.; Galvez-Aranda, D.; Balbuena, P.; Seminario, J., Crystals 2018, 8 (1), 33.