Reactive force-field molecular dynamics study on graphene oxide reinforced cement composite: functional group de-protonation, interfacial bonding and strengthening mechanism
Summary¶
Graphene oxide (GO) is widely explored as a nanoscale reinforcement in cement matrices, but atomistic mechanisms coupling GO functional groups to calcium–silicate–hydrate (C–S–H) gels remain under-characterized. This PCCP article uses ReaxFF MD via LAMMPS (reax/c) to model C–S–H gel with an interlayer insert of graphene or GO decorated with –OH, epoxy, –SO\(_3\)H, and –COOH groups at ~10% coverage (per the manuscript). The work tracks water dissociation, proton transfer, Ca–O\(_c\)–Ca bridge formation, and uniaxial tension responses, ranking GO–COOH and GO–OH composites highest in interfacial strength and ductility relative to pristine graphene or epoxide/sulfonate-dominated cases.
Methods¶
The substrate is a tobermorite 11 Å-like C–S–H cell cleaved to introduce an ~10 Å channel into which graphene/GO sheets are placed. Reactive MD uses LAMMPS with ReaxFF and 3D periodic boundary conditions (PBC) on supercells reaching ~63k–72k atoms for mechanical replicas. Verlet integration uses 0.25 fs timestep; NPT equilibration 250 ps at 300 K and 1 atm with Nosé–Hoover thermostat/barostat, followed by NVT 100 ps and extended 1000 ps production relaxation to allow chemistry. Uniaxial strain is applied at 0.08 ps⁻¹ along interlayer c under NPT with a,b stresses coupled toward approximate zero lateral stress. N/A — external electric field or umbrella sampling in the protocol summarized here.
Findings¶
Outcomes & mechanisms: Deprotonation propensity near C–S–H follows COOH (≈SO\(_3\)H) > OH > epoxy in the scenarios reported; COO⁻–Ca bridges and H-bonds to Si–OH strengthen GO–COOH interfaces, while epoxide opens only ~8% under Ca, leaving GO–epoxide interfaces comparatively weak.
Comparisons: the authors relate interfacial cohesive response to prior GO–cement experiments in the Introduction (e.g. Lv, Pan, Lu citations in the article) as motivation; atomistic tensile metrics should be read as nanoscale benchmarks, not direct macroscopic strength predictions.
Sensitivity: functional group identity and coverage (~10% in the models) dominate interface chemistry; strain rate (0.08 ps⁻¹) is high relative to laboratory tests, so ductility rankings are most meaningful comparatively within this ReaxFF study.
Limitations / outlook: idealized C–S–H morphology and ReaxFF transferability caveats are stated in the article; mesoscale porosity and aggregate microstructure are outside the slit-pore supercells.
Corpus honesty: quantitative stress values and figure references belong in the PCCP PDF (pdf_path); verify before quoting in benchmarks.
Limitations¶
ReaxFF transferability; idealized C–S–H morphology; high strain rate in mechanical tests compared to experiments.
Reader notes (MAS / retrieval)¶
Route GO–cement interface questions here when the user names deprotonation order, Ca bridges, or uniaxial tension of nanocarbon in a C–S–H slit pore.
Cross-link theme-oxides-silica-ceramics for C–S–H background.
Relevance to group¶
Cement–nanocarbon ReaxFF interface chemistry adjacent to oxide surface reactivity studies in the knowledge base.
Citations and evidence anchors¶
- DOI:
10.1039/c8cp00006a.