Supporting Information: Simulation Protocol for Prediction of a Solid-Electrolyte Interphase on Silicon-Based Anodes (ReaxFF)

Summary¶

This corpus PDF is supporting information for a ReaxFF study of solid-electrolyte interphase (SEI) formation on Si / SiO\(_x\) anodes in lithium-ion cells, extending an earlier Si–Li–O reactive parameterization toward Si–Li–O–C–H–F chemistry to represent carbonate electrolytes and additives such as fluorinated species. The SI documents DFT training data choices, ReaxFF optimization scope for additional bond/angle/torsion terms, and simulation protocol notes tied to the parent letter, enabling reproducibility checks for the parameter extension without duplicating the main text narrative. For corpus-wide notes on SI/galley/proof roles, see NON_PRIMARY_ARTICLE_PAPER_SLUGS.md.

Methods¶

Force-field extension (SI). Yun et al. extend prior Si–Li–O ReaxFF parameters (Refs. S1/S2) with literature C–H and O–H blocks (Refs. S3/S4), then develop and optimize new bond, angle, and torsion terms for Si–C, Li–C, O–C, Si–H, Li–H, F–F, C–F, H–F, Li–F, Si–F to reach a Si–Li–O–C–H–F training space suited to carbonate electrolytes and F-containing additives (papers/ReaxFF_others/Kang_Seop_Yun_SiC_EC_JPC_Lett_2017_SI.pdf). QM reference data come from Q-Chem B3LYP/6-311G** scans of bond dissociation, angle bending, and torsions on molecular training sets, plus VASP PBE PAW plane-wave EOS work on SiC crystals (400 eV cutoff, Monkhorst–Pack k-grids). EC reduction pathways and an EC–vinylene carbonate (VC) complex supply additional DFT targets called out on SI pp. S1–S2. Optimizer software (CMA-ES vs. least-squares): N/A — not stated on the indexed SI pages used here.

Production MD for SEI morphology. The SI states that large-scale ReaxFF molecular dynamics will probe SEI formation on Si / SiO\(_x\), but the indexed SI pages used here do not spell out supercell size, PBC, timestep, thermostat/barostat, temperature programs, pressure, electric fields, or production duration—those details belong to the parent J. Phys. Chem. Lett. article, not this SI excerpt alone. Ensemble (NVT vs. NPT) for those production trajectories is not stated on the indexed SI pages (N/A — see parent article).

Standalone static QM discovery: N/A — DFT entries are training data for ReaxFF, not separate materials predictions in this file.

Findings¶

This SI documents which QM surfaces anchor the new Si–C / Li–C / O–C / F interactions and how EC/VC chemistry constrains carbonate reduction energetics; table residuals between B3LYP/PBE and the fitted ReaxFF energies are the right place to judge fit quality. Morphology, cycling, and experimental SEI validation remain in the parent letter—do not cite this PDF alone for those results.

Limitations¶

Wiki metadata here tracks the SI PDF; for full scientific conclusions and validation against experiments, pair this page with the parent J. Phys. Chem. Lett. article PDF/metadata in the corpus.

Relevance to group¶

ReaxFF parameterization workflow for battery interphases on silicon anodes—adjacent to group interests in reactive MD for electrochemical interfaces.

Citations and evidence anchors¶

Treat as SI companion to the KIST/Sejong SiC/electrolyte ReaxFF letter; confirm parent DOI from your VOR PDF when available.

Use this page together with the parent J. Phys. Chem. Lett. article for quantitative SEI claims; the SI PDF records QM training curves and parameter ranges that main-text abstracts typically omit. When citing EC/VC reaction energetics, point readers to tables in the SI and the letter figures for validation against experiment. Charge equilibration schedules and cutoff choices for QM training geometries appear in the SI narrative and should be mirrored if re-fitting parameters.