Skip to content

Reactive molecular simulation of the damage mitigation efficacy of POSS-, graphene-, and carbon nanotube-loaded polyimide coatings exposed to atomic oxygen bombardment

Reactive MD compares atomic oxygen bombardment of polyimide nanocomposites loaded with pristine vs PI-grafted POSS, graphene, and CNTs, emphasizing nanoparticle orientation and exposed PI at the surface.

Summary

Reactive molecular dynamics (methodology adapted from Rahnamoun & van Duin) models hyperthermal AO impacts on PI with POSS, graphene, or CNT fillers at controlled wt % and alignment. Observables include mass loss, erosion yield, surface damage, AO penetration depth, and temperature rise. Random Gr/CNT and grafted POSS generally outperform pristine POSS and aligned Gr/CNT at equal loading. Exposed PI (highest intrinsic erosion yield) dominates composite erosion when nanoparticles remain undamaged.

Methods

A — Force-field training / fitting: Reactive MD follows methodology adapted from Rahnamoun & van Duin—C/H/O/N ReaxFF chemistry for polyimide and nanofillers without new GA/QM training details on this page (see Computational details in the article).

B — Molecular dynamics / reactive sampling: Hyperthermal atomic oxygen bombardment of neat PI and composites: pristine vs PI-grafted POSS (15%, 30%), random vs aligned graphene and CNT (15%); simulation cells ~43–46 Å (Table 1). Tracks mass loss, erosion yield, damage, AO penetration, temperature rise; RDF compares POSS dispersion.

C — DFT / static QM: Not stated as the primary layer in the summarized protocol—ReaxFF drives reactive impact simulations.

D — Review / non-simulation framing: Application study for space-environment polyimide erosion; not a review.

Engine: LAMMPS ReaxFF reactive MD for hyperthermal atomic oxygen impacts (per Computational details in ACS Appl. Mater. Interfaces). System: neat polyimide and nanocomposite slabs / supercells with POSS (15%, 30%), graphene, and CNT (15%); lateral box sizes about 43–46 Å (Table 1); atom totals and stoichiometry are tabulated in the PDF. Ensemble: NVT is the usual impact setup in this literature line; confirm the authors’ declared ensemble in Computational details. Timestep / thermostat / duration / PBC / barostat: N/A — not transcribed on this wiki page—copy from pdf_path. Temperature: hot AO beam conditions and thermostat settings are defined in the article’s impact protocol. Pressure / gas: AO flux and beam energy are specified in-source (not duplicated here). Electric field: N/A — not used. Replica / enhanced sampling: N/A — not used.

Findings

  • Random Gr/CNT and PI-g-POSS show lower mass loss, erosion, damage, penetration, and heating than pristine POSS or aligned Gr/CNT at the same nanoparticle concentration.
  • Exposed PI area on the surface is the dominant lever for erosion yield while nanoparticles remain intact—consistent with early-stage experimental degradation data cited in the paper.
  • PI-grafted POSS achieves much lower erosion than aligned Gr/CNT systems because less PI is exposed; grafting vs pristine POSS lowers erosion by ~4× (low conc.) and increasing POSS loading helps (~1.5×) via better dispersion (RDF evidence).
  • Aligned nanoparticles leave more PI uncovered, raising erosion vs random orientation.

Limitations

Idealized PI monomer representation and specific AO flux/energy window; extrapolation to space mission environments needs experimental validation beyond cited early data. The study’s conclusions are inherently morphology- and orientation-dependent (random vs aligned fillers), so quantitative erosion yield comparisons require care when translating to different nanocomposite processing histories.

Relevance to group

LEO AO + polyimide + ReaxFF damage study—close to prior van Duin-group AO/polymer reactive MD cited in the introduction. Use this page when benchmarking nanocomposite erosion against neat polymer baselines under hyperthermal oxygen conditions relevant to low Earth orbit exposure models.

Citations and evidence anchors

  • DOI: 10.1021/acsami.7b02032
  • Full article tables list cell sizes and nanoparticle loadings referenced in the Methods summary above.