Skip to content

Revealing graphene oxide toxicity mechanisms: a reactive molecular dynamics study

Summary

Golkaram and van Duin report ReaxFF-based reactive MD simulations aimed at atomistic mechanisms by which graphene oxide (GO) functional groups interact with a model peptide helix, interpreted as proxies for biocompatibility/toxicity-relevant chemistry (DOI 10.1016/j.md.2015.10.001). The study progresses from isolated functional-group motifs (epoxide, hydroxyl, carboxyl) to a combined GO model, then examines adhesion, secondary-structure disruption, and solution acidification metrics discussed in the article.

Methods

MD application (atomistic dynamics)

Simulations use ReaxFF reactive MD (article text). A 12-residue α-helix serves as a compact peptide proxy, with a mutated sequence (A…W → …G substitution in Methods) to show trends are not sequence-unique. Graphene sheets in a ~29.56 Å × 34.16 Å periodic supercell host oxygenated motifs at C₉₆O₇, C₉₆(OH)₇, C₉₆(COOH)₇, and combined layouts; explicit water plus peptide yields thousands of atoms in the largest cells (exact atom counts in the PDF). Peptides start ~3 Å from the surface. After 1 K minimization and equilibration (Methods), production NVT runs use Verlet integration, 0.1 fs timestep, 200 ps segments, and a Berendsen thermostat (100 fs damping) per the uncorrected proof text. Engine (LAMMPS vs ADF): N/A — not spelled out in the proof excerpt used here; confirm in the version-of-record PDF if package identification is required. External stress loading / hydrostatic pressure: N/A — not part of this constant-volume peptide–GO protocol.

Force-field training

N/A — applies ReaxFF; parameter lineage is described via citations in the article rather than as a new parameterization performed in this manuscript.

Static QM / DFT

N/A for the main MD trajectories — the discussion cites DFT mainly as supporting context for H-bond energetics comparisons where noted in the article.

Findings

Epoxide motifs can promote water attack that yields hydroxyl pairs on the sheet; the narrative links such chemistry to ROS-prone scenarios discussed relative to biophysical oxidation literature. Carboxyl-rich patches can acidify the local environment, protonate acidic residues such as Asp, and engage Cys thiol chemistry (including disulfide-related disruption) as illustrated in figures. The combined GO model discusses competition or cooperation between epoxide and carboxyl groups in adhesion, loss of secondary structure, and pH drift metrics reported in the article. Sensitivity to functional-group density is intentionally high to magnify signals—a limitation for quantitative dosimetry versus polydisperse experiment. Corpus honesty: this repo’s proof PDF may differ from the final Materials Discovery layout; cite the PDF for numbers.

Limitations

The local corpus file is an uncorrected proof PDF; prefer the final Materials Discovery PDF for pagination/wording. Functional-group densities are intentionally high/exaggerated to magnify signals, so quantitative dosimetry should not be extrapolated linearly to dilute, polydisperse GO samples without additional modeling.

Relevance to group

Adri C. T. van Duin co-authors a reactive MD contribution on graphene oxide biointerface chemistry in Materials Discovery, consistent with the group’s broader nanocarbon and ReaxFF application portfolio.

Citations and evidence anchors