Skip to content

Supporting information: ReaxFF short-range repulsion training for noble gas ion irradiation of graphene

Supporting information PDF

This slug tracks the ACS Nano SI PDF (papers/Yoon_ACSNano_SI.pdf). The peer-reviewed article + DOI live on [[2016yoon-venue-nn6b03036]].

Summary

The Supporting Information for Yoon et al., ACS Nano 2016 documents how short-range noble-gas-ion / carbon repulsion was added to the graphene ReaxFF description used in the main irradiation study. DFT benchmarks use benzene + noble-gas ion clusters as a planar aromatic proxy for graphene, comparing energies to ZBL universal repulsion and to the fitted ReaxFF terms. B3LYP with 6-311G** bases is used for He / Ne / Ar complexes and LACV3P** for Kr (stated in §S1). Three impact sites—ring center, C–C bond center, and atop C—anchor the training geometries referenced in Figure S1.

Methods

1 — MD application (production irradiation MD). §S1 of this SI focuses on QM training of short-range ion–carbon repulsion; the LAMMPS ReaxFF irradiation workflow is published in [[2016yoon-venue-nn6b03036]]. The following bullets summarize that VOR protocol so readers do not have to cross-open files for the headline MD controls: Engine / code: LAMMPS with ReaxFF (C-2013-class parameters plus the SI repulsion extension). System size & composition: periodic graphene supercells with in-plane footprint ~52 × 40 Å\(^2\) plus He\(^+\), Ne\(^+\), Ar\(^+\), or Kr\(^+\) projectiles; exact atom counts follow the VOR Methods text. Boundaries / periodicity: in-plane periodic graphene with thermostatted edge regions acting as heat baths. Ensemble: NVE during each impact cascade; Nosé–Hoover coupling on edge atoms between impacts. Timestep: 0.005–0.02 fs during collisions. Duration / stages: high effective dose rates with ps-scale collision segments, followed by 25 ps anneals at 1500 K and longer 1.25 ns anneals at 2000 K for selected cases (VOR). Thermostat: Nosé–Hoover on edge regions. Barostat / hydrostatic pressure control: N/A — not NPT. Temperature: 300 K pre-equilibration of the sheet; 1500 K/2000 K post-irradiation anneals as quoted above. Pressure: N/A — no external stress target. Electric field: N/A — not applied. Replica / enhanced sampling: N/A — not used.

2 — Force-field training (ReaxFF ion–C repulsion). Goal: augment the C-2013-class ReaxFF description so high-energy noble-gas impacts retain accurate nuclear repulsion without spoiling equilibrium graphene chemistry. QM training data / training set: DFT interaction energies for noble-gas ion + benzene clusters at the B3LYP/6-311G** level (LACV3P** for Kr), compared against ZBL reference data curves (§S1). Geometry coverage: He, Ne, Ar, and Kr ions approached at the three impact sites enumerated above. Optimization: iterative ReaxFF parameter adjustments to minimize disagreement with the DFT/ZBL training curves (details + parameter tables reside in the SI PDF).

3 — Static QM beyond training scans. N/A — §S1 focuses on the repulsive training surfaces only.

4 — Experiments. N/A — STEM/He-ion comparisons live in the main article.

Findings

Outcomes / mechanism (parameter layer). Figure S1 documents that DFT, ZBL, and the updated ReaxFF short-range terms track each other for the benzene + ion relative energy curves at the three impact sites sampled.

Comparisons. Training explicitly benchmarks against DFT and universal ZBL repulsion rather than against experiment at this stage.

Sensitivity / levers. Ion species (He/Ne/Ar/Kr) and approach site (ring, bond, atop) shift the repulsive energy surfaces that the ReaxFF fit must reproduce.

Limitations / outlook. Benzene is a proxy; final performance must be judged in the periodic graphene cascades reported in [[2016yoon-venue-nn6b03036]], where electronic stopping remains neglected.

Corpus honesty. This slug is SI-only; cite the main article for DOI, dose, timestep, anneal, and experimental comparisons.

Limitations

  • SI fragment — always pair with the VOR article page for DOI, pagination, and experimental comparisons.

Relevance to group

Documents PSU/ORNL ReaxFF extension work underpinning ion-engineered graphene defect simulations in the corpus.

Citations and evidence anchors

  • Companion article: [[2016yoon-venue-nn6b03036]] (papers/Yoon_ACSNano_2016.pdf).
  • SI PDF: papers/Yoon_ACSNano_SI.pdf.