Skip to content

Salt concentration effects on mechanical properties of LiPF6/poly(propylene glycol) diacrylate solid electrolyte: Insights from reactive molecular dynamics simulations

Summary

Reactive MD (ReaxFF) study of mechanical properties of LiPF\(_6\)-doped poly(propylene glycol) diacrylate (PPGDA) solid polymer electrolyte motifs aimed at structural battery coatings. The work targets how salt concentration shifts stiffness, failure under multiaxial loading scenarios, and contrasts isotropic compression vs expansion and shear responses, including hydrostatic failure behavior.

The introduction positions PPGDA as a room-temperature-processable coating whose conductivity near 10\(^{-6}\) S/cm motivates ultrathin conformal layers in structural-battery architectures, and frames reactive MD as a way to probe how LiPF\(_6\) loading couples to load-bearing capacity under hydrostatic, expansion, shear, and compression modes.

Methods

1 — MD application (ReaxFF in LAMMPS). Engine: LAMMPS with ReaxFF plus QEq charge equilibration in USER-REAXC, using the C/H/P/F/O/S/Li parameter set previously applied to battery-related systems (§2.2). Boundary conditions: three-dimensional periodic PPGDA supercells as built in §2.1. Structures: crosslinked and non-crosslinked PPGDA cells built with Molden then assembled in LAMMPS (§2.1); orthogonal ~520-atom cells (~17 Å × ~18 Å in-plane averages at 300 K, 1 atm) doped with LiPF\(_6\) at Li:EO = 1:16 and 1:32 (EO = ethylene-oxide atoms in the repeat), with 2×2×2 supercells for finite-size checks (§2.1). Minimization: conjugate-gradient to \(10^{-6}\) kcal/mol/Å max force with quadratic line search (\(10^{-2}\) Å max displacement). Integration: Verlet with 0.25 fs timestep. Equilibration: NPT at 300 K, 1 atm using Nosé–Hoover chains (five thermostats) with 100 fs temperature damping and 250 fs pressure damping (anisotropic stress control as stated). Follow-on segments: NVT Nosé–Hoover equilibration until mean potential-energy drift is \(\mathcal{O}(1)\) kcal/mol/ns (§2.2 narrative). Mechanical tests: dynamic tensile stiffness scans along backbone-orthogonal directions under NPT with alternative damping tuples (100/250, 25/25, 25/250 fs); strain rates in [0.08, 8] ns\(^{-1}\) for stiffness; simple shear failure at 12.5 ns\(^{-1}\); Müller–Plathe setup for non-equilibrium shear viscosity (§2.2). Electric field: N/A — not used. Replica / enhanced sampling: N/A — not used.

2 — Force-field training. N/A — the study employs a published ReaxFF database; §3.1 compares selected bond-dissociation and charge metrics against DFT rather than refitting here.

3 — Static QM (validation DFT). VASP PAW GGA-PBE with DFT-D3(BJ) dispersion for condensed motifs; electronic convergence and ionic relaxation criteria, k-point (Monkhorst–Pack) settings, and 400/520 eV cutoffs for fixed/variable cell tasks as listed in §2.2. Atomic charges: DDEC post-processing for comparison to ReaxFF/QEq charges.

4 — Continuum or mesoscale. N/A — atomistic RMD focus.

Findings

Mechanical trends (abstract + §3). Stiffness can increase beyond a threshold LiPF\(_6\) concentration, while failure behavior depends on loading path. Isotropic expansion and shear show modest decreases in failure strength and failure strain as salt loading rises. Isotropic hydrostatic compression produces no bond-dissociation failure up to 10 GPa in the reported simulations, contrasting with expansion/shear channels.

Validation / transferability. §3.1 documents ReaxFF vs DFT gaps for ether dissociation, LiPF\(_6\) dissociation energies, and PPGDA crosslink / monomer energetics (Table 1), plus PEO melting point, density, viscosity, and LiPF\(_6\)/PEO transport benchmarks (Tables 2–3).

Sensitivity / levers. Salt stoichiometry, crosslink density, and multiaxial loading direction change which C–O acrylate bonds break first, linking microscopic bond statistics to macroscopic stiffness and failure trends discussed in §3.4.

Limitations (authored). Abstract positions ~10\(^{-6}\) S/cm room-temperature conductivity as an acceptable trade-off for ultrathin structural-battery coatings; ReaxFF barriers may underestimate absolute failure stresses relative to DFT (§3.1 discussion).

Limitations

  • ReaxFF accuracy for salt-polymer chemistries should be checked case-by-case; §3.1 flags systematic gaps vs DFT for some dissociation pathways.
  • Room-temperature conductivity (~10\(^{-6}\) S/cm class in Introduction) is low; structural-battery use cases assume architectures tolerating that scale (as discussed in the paper).

Relevance to group

van Duin-group coauthored ReaxFF investigation of polymer electrolyte mechanics—pairs naturally with other battery interface reactive MD notes in the wiki.

Citations and evidence anchors

  • DOI: 10.1016/j.electacta.2016.10.035
  • Text-aligned pointers: normalized/extracts/2016verners-electrochimi-salt-concentration_p1-2.txt (truncated; use PDF for full Methods)