Skip to content

Supporting information: atomistic observation of the lithiation and delithiation behaviors of silicon nanowires using reactive molecular dynamics simulation

Evidence and attribution

Authority of statements

This slug registers supporting information PDF bytes. Narrative claims about the main study should be checked against the primary article PDF, not SI alone.

Summary

SI documents development of a Li–Si ReaxFF parametrization: DFT training on Li₄Si clusters and several LiₓSi crystals; B3LYP/6-311G** cluster data in Q-Chem; periodic PBE (VASP) equations of state and Bader charges for charge-fitting; successive one-parameter search optimization. Parameters build on prior Si and Li ReaxFF sets with new Li–Si bond terms (extract section S1).

Methods

Section (S1) documents ReaxFF optimization for Li–Si against a DFT training set: cluster data for a Li\(_4\)Si molecule (B3LYP/6-311G in Q-Chem) and periodic PBE (VASP) equations of state and Bader charges for several thermodynamically stable Li\(_x\)Si crystals (Li\(_{21}\)Si\(_5\), Li\(_{15}\)Si\(_4\), Li\(_{13}\)Si\(_4\), Li\(_7\)Si\(_3\), Li\(_{12}\)Si\(_7\), LiSi). Plane-wave PBE calculations use a 400 eV cutoff and 11×11×11 Monkhorst–Pack k-mesh for energy convergence. Coulomb-related ReaxFF terms are fitted using Bader charges from the DFT densities. Optimization uses a successive one-parameter search. The Li–Si development starts from existing Si (van Duin et al.) and Li (Han et al.) ReaxFF subsets, but Li–Si bond parameters are newly derived for this work; parameters appear in Table S1 (SI pages S1–S2; extract).

1 — MD application (parent-article nanowire simulations)

  • Engine / code: N/A in this SI excerpt; the parent article’s Si nanowire lithiation ReaxFF trajectories are described on [[2015han-venue-atomistic-observation]] using LAMMPS molecular dynamics workflows (papers/ReaxFF_others/Han_LiSi_JPC_2015.pdf).
  • System size & composition: N/A in SI §S1 for nanowire atom counts; SI focuses on Li–Si QM training cells (Li\(_4\)Si cluster and Li\(_x\)Si crystals) containing tens to hundreds of atoms per formula unit stack (extract).
  • Boundaries / periodicity: 3D PBC for VASP PBE EOS grids on Li\(_x\)Si crystals (extract).
  • Ensemble (production MD): N/A — not specified in SI §S1; see parent article for NVT/NPT choices.
  • Timestep / duration / thermostat (production MD): N/A — not specified in SI §S1; see parent article for fs timestep and ps/ns segments.
  • Barostat / pressure (production MD): N/A — hydrostatic pressure control is not described in SI §S1; confirm in parent article.
  • Temperature (production MD): N/A — not described in SI §S1; parent article lists operating temperatures.
  • DFT reference temperature: PBE EOS energies are 0 K DFT total energies on k-meshed Brillouin-zone samples (extract).

2 — Force-field training (Li–Si ReaxFF extension)

  • Parent FF / elements: extends prior Si (van Duin et al.) and Li (Han et al.) ReaxFF subsets with new Li–Si bond parameters (SI §S1).
  • QM reference (clusters): B3LYP/6-311G in Q-Chem on Li\(_4\)Si (extract).
  • QM reference (periodic): PBE plane-wave VASP EOS and Bader charges for listed Li\(_x\)Si crystals; 400 eV cutoff, 11×11×11 Monkhorst–Pack k-mesh (extract).
  • Training observables: EOS energies/charges for crystals; cluster data for Li–Si bonding (extract).
  • Optimization: successive one-parameter search updating ReaxFF parameters (extract).
  • Tables: Table S1 lists fitted coefficients (SI).

Findings

1 — Outcomes (SI scope)

The SI states explicitly that Si and Li parameters from prior fits were reused as a base, while terms governing Li–Si bonds were newly trained to support the nanowire lithiation/delithiation reactive MD study in the parent article (SI text; extract pages S1–S2).

2 — Comparisons

  • Bader-charge-guided Coulomb terms are tied to DFT Bader analysis so Li–Si bond-order parameters track QM without refitting the entire Si/Li subsets (extract).

3 — Sensitivity

N/A — not applicable at SI parameter-documentation level.

4 — Limitations / outlook

SI documents QM fits, not full device-scale electrode workflows—see parent article.

5 — Corpus / KB honesty

This page summarizes SI pages S1–S2 only (normalized/extracts/2014han-venue-supporting-information_p1-2.txt); do not treat it as the version-of-record article record.

Limitations

SI is not a standalone experimental or methods paper; figures and tables are auxiliary to the parent article. Nanowire lithiation simulations in the parent work may use additional electrode boundary conditions beyond the parameter development excerpts summarized here; treat this page as QM-fit documentation, not a substitute for the full reactive MD protocol in the main PDF. Table S1 coefficients must be copied verbatim into input decks with consistent units for your ReaxFF software flavor.

Reader notes (navigation)

Primary article PDF in this corpus: 2015han-venue-atomistic-observation (papers/ReaxFF_others/Han_LiSi_JPC_2015.pdf). - Maintainer catalog (SI-first ingests): Non-primary article PDF slugs (GitHub) section A (analogous SI roles).

Citations and evidence anchors

  • normalized/extracts/2014han-venue-supporting-information_p1-2.txt (SI pages S1–S2).